eo-poverty-review
An Overview of Earth Observation in Poverty Research
https://github.com/aiandglobaldevelopmentlab/eo-poverty-review
Science Score: 54.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
✓CITATION.cff file
Found CITATION.cff file -
✓codemeta.json file
Found codemeta.json file -
○.zenodo.json file
-
✓DOI references
Found 102 DOI reference(s) in README -
✓Academic publication links
Links to: arxiv.org -
○Academic email domains
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (8.5%) to scientific vocabulary
Last synced: 6 months ago
·
JSON representation
·
Repository
An Overview of Earth Observation in Poverty Research
Basic Info
- Host: GitHub
- Owner: AIandGlobalDevelopmentLab
- Language: TeX
- Default Branch: main
- Size: 46.9 KB
Statistics
- Stars: 0
- Watchers: 4
- Forks: 0
- Open Issues: 0
- Releases: 0
Created almost 2 years ago
· Last pushed almost 2 years ago
Metadata Files
Readme
Citation
README.md
Awesome Papers on Earth Observation (EO) in Machine Learning (ML) and Causal Inference (CI)
Description: This repository contains a directory of papers on earth observation, causal inference, machine learning, and/or poverty research compiled by Kaz Sakamoto for a scoping literature review in collaboration with Adel Daoud and Connor T. Jerzak. To find associated .bib entries, see eo-poverty-review/citations.
Contribute: You can additional papers for inclusion by opening an issue in this repo (feel free to suggest your own work!).
Overview | Literature Directory | Reference
Paper Directory
Ali, Sahara, Omar Faruque, Yiyi Huang, Md. Osman Gani, Aneesh Subramanian, Nicole-Jeanne Schlegel, and Jianwu Wang. 2023. “Quantifying Causes of Arctic Amplification via Deep Learning Based Time-Series Causal Inference.” In 2023 International Conference on Machine Learning and Applications (ICMLA), 689–96. https://doi.org/10.1109/ICMLA58977.2023.00101.
Berrevoets, Jeroen, Krzysztof Kacprzyk, Zhaozhi Qian, and Mihaela van der Schaar. 2024. “Causal Deep Learning.” arXiv. http://arxiv.org/abs/2303.02186.
Biswas, Mriganka Sekhar, and Manmeet Singh. 2022. “Trustworthy Modelling of Atmospheric Formaldehyde Powered by Deep Learning.” arXiv. http://arxiv.org/abs/2209.07414.
Boussard, Julien, Chandni Nagda, Julia Kaltenborn, Charlotte Emilie Elektra Lange, Philippe Brouillard, Yaniv Gurwicz, Peer Nowack, and David Rolnick. 2023. “Towards Causal Representations of Climate Model Data.” arXiv. http://arxiv.org/abs/2312.02858.
Camps-Valls, G. 2021. “Perspective on Deep Learning for Earth Sciences.” In Generalization With Deep Learning: For Improvement On Sensing Capability, 159–73. https://doi.org/10.1142/9789811218842_0007.
Camps-Valls, G., M. Reichstein, X. Zhu, and D. Tuia. 2020. “ADVANCING DEEP LEARNING for EARTH SCIENCES: From HYBRID MODELING to INTERPRETABILITY.” In, 3979–82. https://doi.org/10.1109/IGARSS39084.2020.9323558.
Camps-Valls, Gustau, Daniel H. Svendsen, Jordi Cortés-Andrés, Álvaro Mareno-Martínez, Adrián Pérez-Suay, Jose Adsuara, Irene Martín, Maria Piles, Jordi Muñoz-Marí, and Luca Martino. 2021. “Physics-Aware Machine Learning for Geosciences and Remote Sensing.” In 2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS, 2086–89. https://doi.org/10.1109/IGARSS47720.2021.9554521.
Chen, Xiangyu, Kaisa Zhang, Gang Chuai, Weidong Gao, Zhiwei Si, Yijian Hou, and Xuewen Liu. 2023. “Urban Area Characterization and Structure Analysis: A Combined Data-Driven Approach by Remote Sensing Information and Spatial-Temporal Wireless Data.” REMOTE SENSING 15 (4): 1041. https://doi.org/10.3390/rs15041041.
Chouzenoux, Emilie, and Victor Elvira. 2023. “Sparse Graphical Linear Dynamical Systems.” arXiv. http://arxiv.org/abs/2307.03210.
Cohrs, Kai-Hendrik, Gherardo Varando, Nuno Carvalhais, Markus Reichstein, and Gustau Camps-Valls. 2024. “Causal Hybrid Modeling with Double Machine Learning.” arXiv. http://arxiv.org/abs/2402.13332.
Das, M., and S. K. Ghosh. 2017. “A Deep-Learning-Based Forecasting Ensemble to Predict Missing Data for Remote Sensing Analysis.” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 10 (12): 5228–36. https://doi.org/10.1109/JSTARS.2017.2760202.
Debeire, Kevin, Jakob Runge, Andreas Gerhardus, and Veronika Eyring. 2024. “Bootstrap Aggregation and Confidence Measures to Improve Time Series Causal Discovery.” arXiv. http://arxiv.org/abs/2306.08946.
Dong, Hongwei, Lingyu Si, Wenwen Qiang, Lamei Zhang, Junzhi Yu, Yuquan Wu, Changwen Zheng, and Fuchun Sun. 2024. “A Novel Causal Inference-Guided Feature Enhancement Framework for PolSAR Image Classification.” IEEE Transactions on Geoscience and Remote Sensing 62: 1–16. https://doi.org/10.1109/TGRS.2023.3343380.
Eldhose, Elizabeth, Tejasvi Chauhan, Vikram Chandel, Subimal Ghosh, and Auroop R Ganguly. n.d. “Robust Causality and False Attribution in Data-Driven Earth Science Discoveries.”
Elvira, Víctor, Émilie Chouzenoux, Jordi Cerdà, and Gustau Camps-Valls. 2023. “Graphs in State-Space Models for Granger Causality in Climate Science.” arXiv. http://arxiv.org/abs/2307.10703.
Fernández-Loría, Carlos, and Jorge Loría. 2024. “Causal Scoring: A Framework for Effect Estimation, Effect Ordering, and Effect Classification.” arXiv. http://arxiv.org/abs/2206.12532.
Giannarakis, Georgios, Ilias Tsoumas, Stelios Neophytides, Christiana Papoutsa, Charalampos Kontoes, and Diofantos Hadjimitsis. 2023. “Understanding the Impacts of Crop Diversification in the Context of Climate Change: A Machine Learning Approach.” arXiv. http://arxiv.org/abs/2307.08617.
Giannarakis, Georgios, Vasileios Sitokonstantinou, Roxanne Suzette Lorilla, and Charalampos Kontoes. 2022. “Towards Assessing Agricultural Land Suitability with Causal Machine Learning.” In 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 1441–51. https://doi.org/10.1109/CVPRW56347.2022.00150.
Giannarakis, G., I. Tsoumas, S. Neophytides, C. Papoutsa, C. Kontoes, and D. Hadjimitsis. 2023. “UNDERSTANDING THE IMPACTS OF CROP DIVERSIFICATION IN THE CONTEXT OF CLIMATE CHANGE: A MACHINE LEARNING APPROACH.” In, 48:1379–84. https://doi.org/10.5194/isprs-archives-XLVIII-1-W2-2023-1379-2023.
Go, Eugenia, Kentaro Nakajima, Yasuyuki Sawada, and Kiyoshi Taniguchi. 2022. “On the Use of Satellite-Based Vehicle Flows Data to Assess Local Economic Activity: The Case of Philippine Cities.” {SSRN} {Scholarly} {Paper}. Rochester, NY. https://doi.org/10.2139/ssrn.4057690.
Gomez-Chova, Luis, Raul Santos-Rodriguez, and Gustau Camps-Valls. 2018. “Signal-to-Noise Ratio in Reproducing Kernel Hilbert Spaces.” PATTERN RECOGNITION LETTERS 112 (September): 75–82. https://doi.org/10.1016/j.patrec.2018.06.004.
Gómez-Chova, Luis, and Gustavo Camps-Valls. 2012. “Learning with the Kernel Signal to Noise Ratio.” In 2012 IEEE International Workshop on Machine Learning for Signal Processing, 1–6. https://doi.org/10.1109/MLSP.2012.6349715.
Harding, M. C., and C. Lamarche. 2021. “Small Steps with Big Data: Using Machine Learning in Energy and Environmental Economics.” Annual Review of Resource Economics 13: 469–88. https://doi.org/10.1146/annurev-resource-100920-034117.
Heuer, Helge, Mierk Schwabe, Pierre Gentine, Marco A. Giorgetta, and Veronika Eyring. 2023. “Interpretable Multiscale Machine Learning-Based Parameterizations of Convection for ICON.” arXiv. http://arxiv.org/abs/2311.03251.
Jay, Jonathan. 2020. “Alcohol Outlets and Firearm Violence: A Place-Based Case-Control Study Using Satellite Imagery and Machine Learning.” INJURY PREVENTION 26 (1): 61–66. https://doi.org/10.1136/injuryprev-2019-043248.
Jerzak, C. T., F. Johansson, and A. Daoud. 2023. “Image-Based Treatment Effect Heterogeneity.” Proceedings of the Second Conference on Causal Learning and Reasoning (CLeaR), Proceedings of Machine Learning Research (PMLR), 213: 531-552. https://www.scopus.com/inward/record.uri?eid=2-s2.0-85147679914&partnerID=40&md5=7866b08ec210422443d7ad823941845f.
Jerzak, Connor T., and Adel Daoud. 2023. “CausalImages: An R Package for Causal Inference with Earth Observation, Bio-Medical, and Social Science Images.” arXiv. http://arxiv.org/abs/2310.00233.
Jerzak, Connor T., Fredrik Johansson, and Adel Daoud. 2023a. “Integrating Earth Observation Data into Causal Inference: Challenges and Opportunities.” arXiv. http://arxiv.org/abs/2301.12985.
Jerzak, Connor T., Fredrik Johansson, and Adel Daoud. 2023b. “Estimating Causal Effects Under Image Confounding Bias with an Application to Poverty in Africa.” arXiv. http://arxiv.org/abs/2206.06410.
Ji, J., T. Wang, J. Liu, M. Wang, and W. Tang. 2024. “River Runoff Causal Discovery with Deep Reinforcement Learning.” Applied Intelligence 54 (4): 3547–65. https://doi.org/10.1007/s10489-024-05348-7.
Li, Hongga, Xiaoxia Huang, and Xia Li. 2019. “Urban Land Price Assessment Based on GIS and Deep Learning.” In IGARSS 2019 - 2019 IEEE International Geoscience and Remote Sensing Symposium, 935–38. https://doi.org/10.1109/IGARSS.2019.8900516.
Li, L. U., W. E. I. Shangguan, Y. I. Deng, J. Mao, J. Pan, N. A. N. Wei, H. U. A. Yuan, S. Zhang, Y. Zhang, and Y. Dai. 2020. “A Causal Inference Model Based on Random Forests to Identify the Effect of Soil Moisture on Precipitation.” Journal of Hydrometeorology 21 (5): 1115–31. https://doi.org/10.1175/JHM-D-19-0209.1.
Li, X., M. Feng, Y. Ran, Y. Su, F. Liu, C. Huang, H. Shen, et al. 2023. “Big Data in Earth System Science and Progress Towards a Digital Twin.” Nature Reviews Earth and Environment. https://doi.org/10.1038/s43017-023-00409-w.
Liang, X. San, X San Liang, Dake Chen, and Renhe Zhang. 2024. “Quantitative Causality, Causality-Guided Scientific Discovery, and Causal Machine Learning.” https://doi.org/10.22541/essoar.170913638.81842156/v1.
Lin, L., L. Di, C. Zhang, L. Guo, H. Zhao, D. Islam, H. Li, Z. Liu, and G. Middleton. 2024. “Modeling Urban Redevelopment: A Novel Approach Using Time-Series Remote Sensing Data and Machine Learning.” Geography and Sustainability 5 (2): 211–19. https://doi.org/10.1016/j.geosus.2024.02.001.
Luo, J., and Q. Zhang. 2019. “Big Data Pioneers New Ways of Geoscience Research: Identifying Relevant Relationships to Enhance Research Feasibility.” Earth Science Frontiers 26 (4): 6–12. https://doi.org/10.13745/j.esf.sf.2019.4.28.
Ma, Qiaoyu, Yang Liu, Xintong Wang, Biao Yuan, and Kai Zhang. 2021. “Hyperspectral Image Recognition Based on Lightweight Causal Convolutional Network.” In 2021 IEEE 2nd International Conference on Big Data, Artificial Intelligence and Internet of Things Engineering (ICBAIE), 431–35. https://doi.org/10.1109/ICBAIE52039.2021.9389851.
Mateo-Sanchis, Anna, Jordi Muñoz-Marí, Adrián Pérez-Suay, and Gustau Camps-Valls. 2020. “Warped Gaussian Processes in Remote Sensing Parameter Estimation and Causal Inference.” arXiv. http://arxiv.org/abs/2012.12105.
Moe, S. Jannicke, John F. Carriger, and Miriam Glendell. 2021. “Increased Use of Bayesian Network Models Has Improved Environmental Risk Assessments.” INTEGRATED ENVIRONMENTAL ASSESSMENT AND MANAGEMENT 17 (1): 53–61. https://doi.org/10.1002/ieam.4369.
Morata-Dolz, Miguel, Diego Bueso, Maria Piles, and Gustau Camps-Valls. 2020. “Understanding Climate Impacts on Vegetation with Gaussian Processes in Granger Causality.” arXiv. http://arxiv.org/abs/2012.03338.
Mu, Bin, Bo Qin, Shijin Yuan, Xin Wang, and Yuxuan Chen. 2023. “PIRT: A Physics-Informed Red Tide Deep Learning Forecast Model Considering Causal-Inferred Predictors Selection.” IEEE Geoscience and Remote Sensing Letters 20: 1–5. https://doi.org/10.1109/LGRS.2023.3250642.
Noy, K., N. Ohana-Levi, N. Panov, M. Silver, and A. Karnieli. 2021. “A Long-Term Spatiotemporal Analysis of Biocrusts Across a Diverse Arid Environment: The Case of the Israeli-Egyptian Sandfield.” Science of the Total Environment 774. https://doi.org/10.1016/j.scitotenv.2021.145154.
Otgonbaatar, Soronzonbold, Mihai Datcu, and Begüm Demir. 2022. “Causality for Remote Sensing: An Exploratory Study.” In IGARSS 2022 - 2022 IEEE International Geoscience and Remote Sensing Symposium, 259–62. https://doi.org/10.1109/IGARSS46834.2022.9883060.
Pérez-Suay, Adrián, and Gustau Camps-Valls. 2019. “Causal Inference in Geoscience and Remote Sensing From Observational Data.” IEEE Transactions on Geoscience and Remote Sensing 57 (3): 1502–13. https://doi.org/10.1109/TGRS.2018.2867002.
Qiu, Xinzhu, Yunzhe Wang, Jingyi Cao, Guannan Xu, Yanan You, and Junlong Ren. 2021. “Economic Development Analysis of the Belt and Road Regions Based on Automatic Interpretation of Remote Sensing Images.” In 2021 7th IEEE International Conference on Network Intelligence and Digital Content (IC-NIDC), 96–101. https://doi.org/10.1109/IC-NIDC54101.2021.9660561.
Ramachandra, Vikas. n.d. “Causal Inference for Climate Change Events from Satellite Image Time Series Using Computer Vision and Deep Learning.”
Ratledge, Nathan, Gabe Cadamuro, Brandon de la Cuesta, Matthieu Stigler, and Marshall Burke. 2022. “Using Machine Learning to Assess the Livelihood Impact of Electricity Access.” NATURE 611 (7936): 491–+. https://doi.org/10.1038/s41586-022-05322-8.
Rong, Yineng, and X. San Liang. 2022. “An Information Flow-Based Sea Surface Height Reconstruction Through Machine Learning.” IEEE Transactions on Geoscience and Remote Sensing 60: 1–9. https://doi.org/10.1109/TGRS.2022.3140398.
Runge, Jakob, Andreas Gerhardus, Gherardo Varando, Veronika Eyring, and Gustau Camps-Valls. 2023. “Causal Inference for Time Series.” NATURE REVIEWS EARTH & ENVIRONMENT 4 (7): 487–505. https://doi.org/10.1038/s43017-023-00431-y.
Runge, J., X.-A. Tibau, M. Bruhns, J. Muñoz-Marí, and G. Camps-Valls. 2019. “The Causality for Climate Competition.” In, 123:110–20. https://www.scopus.com/inward/record.uri?eid=2-s2.0-85102541381&partnerID=40&md5=12adb7bf6e4901f53374a3d818bf7f19.
Sarkodie, S. A. 2020. “Causal Effect of Environmental Factors, Economic Indicators and Domestic Material Consumption Using Frequency Domain Causality Test.” Science of the Total Environment 736. https://doi.org/10.1016/j.scitotenv.2020.139602.
Seong, Nohyoon. 2021. “Deep Spatiotemporal Attention Network for Fine Particle Matter 2.5 Concentration Prediction With Causality Analysis.” IEEE Access 9: 73230–39. https://doi.org/10.1109/ACCESS.2021.3080828.
Serdavaa, Batkhurel. 2023. “A Satellite Image Analysis on Housing Conditions and the Effectiveness of the Affordable Housing Mortgage Program in Mongolia: A Deep Learning Approach.” {SSRN} {Scholarly} {Paper}. Rochester, NY. https://doi.org/10.2139/ssrn.4664966.
Sharma, Somya, Swati Sharma, Rafael Padilha, Emre Kiciman, and Ranveer Chandra. 2024. “Domain Adaptation for Sustainable Soil Management Using Causal and Contrastive Constraint Minimization.” arXiv. http://arxiv.org/abs/2401.07175.
Su, J., D. Chen, D. Zheng, Y. Su, and X. Li. 2023. “The Insight of Why: Causal Inference in Earth System Science.” Science China Earth Sciences 66 (10): 2169–86. https://doi.org/10.1007/s11430-023-1148-7.
Tan, S.-C., G.-Y. Shi, J.-H. Shi, H.-W. Gao, and X. Yao. 2011. “Correlation of Asian Dust with Chlorophyll and Primary Productivity in the Coastal Seas of China During the Period from 1998 to 2008.” Journal of Geophysical Research: Biogeosciences 116 (2). https://doi.org/10.1029/2010JG001456.
Tesch, T., S. Kollet, and J. Garcke. 2023. “Causal Deep Learning Models for Studying the Earth System.” Geoscientific Model Development 16 (8): 2149–66. https://doi.org/10.5194/gmd-16-2149-2023.
Tesch, Tobias, Stefan Kollet, and Jochen Garcke. 2021. “Variant Approach for Identifying Spurious Relations That Deep Learning Models Learn.” FRONTIERS IN WATER 3 (September): 745563. https://doi.org/10.3389/frwa.2021.745563.
Tuia, Devis, Ribana Roscher, Jan Dirk Wegner, Nathan Jacobs, Xiaoxiang Zhu, and Gustau Camps-Valls. 2021. “Toward a Collective Agenda on AI for Earth Science Data Analysis.” IEEE Geoscience and Remote Sensing Magazine 9 (2): 88–104. https://doi.org/10.1109/MGRS.2020.3043504.
Wang, Chenguang, Yepeng Liu, Xiaojian Zhang, Xuechun Li, Vladimir Paramygin, Arthriya Subgranon, Peter Sheng, Xilei Zhao, and Susu Xu. 2023. “Causality-Informed Rapid Post-Hurricane Building Damage Detection in Large Scale from InSAR Imagery.” arXiv. http://arxiv.org/abs/2310.01565.
Wang, Guangxing, Guoshuai Dong, Hui Li, Lirong Han, Xuanwen Tao, and Peng Ren. 2019. “Remote Sensing Image Synthesis via Graphical Generative Adversarial Networks.” In IGARSS 2019 - 2019 IEEE International Geoscience and Remote Sensing Symposium, 10027–30. https://doi.org/10.1109/IGARSS.2019.8898915.
Wang, Jun, and Klaus Mueller. 2016. “The Visual Causality Analyst: An Interactive Interface for Causal Reasoning.” IEEE Transactions on Visualization and Computer Graphics 22 (1): 230–39. https://doi.org/10.1109/TVCG.2015.2467931.
Wang, Yu-Hsiang, You Cartus Bo-Xiang, Hsiung-Ming Liao, Ming-Ching Chang, and Richard Tsai. 2024. “Deeppvmap: Deep Photovoltaic Map for Efficient Segmentation of Solar Panels from Low-Resolution Aerial Imagery.” {SSRN} {Scholarly} {Paper}. Rochester, NY. https://doi.org/10.2139/ssrn.4779346.
Wei, C.-C. 2014. “Meta-Heuristic Bayesian Networks Retrieval Combined Polarization Corrected Temperature and Scattering Index for Precipitations.” Neurocomputing 136: 71–81. https://doi.org/10.1016/j.neucom.2014.01.030.
Weichwald, Sebastian, Martin E. Jakobsen, Phillip B. Mogensen, Lasse Petersen, Nikolaj Thams, and Gherardo Varando. 2020. “Causal Structure Learning from Time Series: Large Regression Coefficients May Predict Causal Links Better in Practice Than Small p-Values.” arXiv. http://arxiv.org/abs/2002.09573.
Xiong, Wei, Zhenyu Xiong, and Yaqi Cui. 2022a. “A Confounder-Free Fusion Network for Aerial Image Scene Feature Representation.” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 15: 5440–54. https://doi.org/10.1109/JSTARS.2022.3189052.
Xiong, Wei, Zhenyu Xiong, and Yaqi Cui. 2022b. “An Explainable Attention Network for Fine-Grained Ship Classification Using Remote-Sensing Images.” IEEE Transactions on Geoscience and Remote Sensing 60: 1–14. https://doi.org/10.1109/TGRS.2022.3162195.
Xu, Fangcao, Jian Sun, Guido Cervone, and Mark Salvador. 2022. “Ill-Posed Surface Emissivity Retrieval from Multi-Geometry Hyperspectral Images Using a Hybrid Deep Neural Network.” arXiv. http://arxiv.org/abs/2107.04631.
Xu, Qingsong, Yilei Shi, Jonathan Bamber, Ye Tuo, Ralf Ludwig, and Xiao Xiang Zhu. 2024. “Physics-Aware Machine Learning Revolutionizes Scientific Paradigm for Machine Learning and Process-Based Hydrology.” arXiv. http://arxiv.org/abs/2310.05227.
Xu, Zhengyi, Wen Jiang, and Jie Geng. 2024. “Texture-Aware Causal Feature Extraction Network for Multimodal Remote Sensing Data Classification.” IEEE Transactions on Geoscience and Remote Sensing 62: 1–12. https://doi.org/10.1109/TGRS.2024.3368091.
Yu, Qiuyan, Wenjie Ji, Lara Prihodko, C. Wade Ross, Julius Y. Anchang, and Niall P. Hanan. 2021. “Study Becomes Insight: Ecological Learning from Machine Learning.” METHODS IN ECOLOGY AND EVOLUTION 12 (11): 2117–28. https://doi.org/10.1111/2041-210X.13686.
Yuan, K., Q. Zhu, W. J. Riley, F. Li, and H. Wu. 2022. “Understanding and Reducing the Uncertainties of Land Surface Energy Flux Partitioning Within CMIP6 Land Models.” Agricultural and Forest Meteorology 319. https://doi.org/10.1016/j.agrformet.2022.108920.
Zhang, Zhaoxiang, Yuelei Xu, Qi Cui, Qing Zhou, and Linhua Ma. 2022. “Unsupervised SAR and Optical Image Matching Using Siamese Domain Adaptation.” IEEE Transactions on Geoscience and Remote Sensing 60: 1–16. https://doi.org/10.1109/TGRS.2022.3170316.
Zheng, Y., H. Zheng, and X. Ye. 2016. “Using Machine Learning in Environmental Tax Reform Assessment for Sustainable Development: A Case Study of Hubei Province, China.” Sustainability (Switzerland) 8 (11). https://doi.org/10.3390/su8111124.
Zheng, Zhuo, Yanfei Zhong, Shiqi Tian, Ailong Ma, and Liangpei Zhang. 2022. “ChangeMask: Deep Multi-Task Encoder-Transformer-Decoder Architecture for Semantic Change Detection.” ISPRS Journal of Photogrammetry and Remote Sensing 183 (January): 228–39. https://doi.org/10.1016/j.isprsjprs.2021.10.015.
Review Reference
Kazuki Sakamoto, Connor T. Jerzak, Adel Daoud. Planetary Causal Inference: Implications for the Geography of Poverty. ArXiv Preprint, 2024. [PDF]
@incollection{sakamoto2024scoping,
author = {Sakamoto, Kazuki and Jerzak, Connor T. and Daoud, Adel},
title = {A Scoping Review of Earth Observation and Machine Learning for Causal Inference: Implications for the Geography of Poverty},
booktitle = {Geography of Poverty},
editor = {Hall, Ola and Wahab, Ibrahim},
year = {2025}
}
Owner
- Name: AIandGlobalDevelopmentLab
- Login: AIandGlobalDevelopmentLab
- Kind: organization
- Repositories: 1
- Profile: https://github.com/AIandGlobalDevelopmentLab
Citation (citations/EOML_Causal.bib)
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% published papers
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
@article{lin_modeling_2024,
title = {Modeling urban redevelopment: {A} novel approach using time-series remote sensing data and machine learning},
volume = {5},
issn = {2096-7438},
shorttitle = {Modeling urban redevelopment},
url = {https://www.sciencedirect.com/science/article/pii/S2666683924000087},
doi = {10.1016/j.geosus.2024.02.001},
abstract = {Accurate mapping and timely monitoring of urban redevelopment are pivotal for urban studies and decision-makers to foster sustainable urban development. Traditional mapping methods heavily depend on field surveys and subjective questionnaires, yielding less objective, reliable, and timely data. Recent advancements in Geographic Information Systems (GIS) and remote-sensing technologies have improved the identification and mapping of urban redevelopment through quantitative analysis using satellite-based observations. Nonetheless, challenges persist, particularly concerning accuracy and significant temporal delays. This study introduces a novel approach to modeling urban redevelopment, leveraging machine learning algorithms and remote-sensing data. This methodology can facilitate the accurate and timely identification of urban redevelopment activities. The study's machine learning model can analyze time-series remote-sensing data to identify spatio-temporal and spectral patterns related to urban redevelopment. The model is thoroughly evaluated, and the results indicate that it can accurately capture the time-series patterns of urban redevelopment. This research's findings are useful for evaluating urban demographic and economic changes, informing policymaking and urban planning, and contributing to sustainable urban development. The model can also serve as a foundation for future research on early-stage urban redevelopment detection and evaluation of the causes and impacts of urban redevelopment. © 2024},
language = {English},
number = {2},
journal = {Geography and Sustainability},
author = {Lin, L. and Di, L. and Zhang, C. and Guo, L. and Zhao, H. and Islam, D. and Li, H. and Liu, Z. and Middleton, G.},
year = {2024},
keywords = {Machine learning, Remote sensing, Time-series analysis, Urban redevelopment, Urban sustainability},
pages = {211--219},
file = {Lin et al. - 2024 - Modeling urban redevelopment A novel approach usi.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\E8R65SMF\\Lin et al. - 2024 - Modeling urban redevelopment A novel approach usi.pdf:application/pdf},
}
@incollection{camps-valls_perspective_2021,
title = {Perspective on deep learning for earth sciences},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85111438128&doi=10.1142%2f9789811218842_0007&partnerID=40&md5=00f3b45c6963e8a52d3bcf0958c63c74},
booktitle = {Generalization {With} {Deep} {Learning}: {For} {Improvement} {On} {Sensing} {Capability}},
author = {Camps-Valls, G.},
year = {2021},
doi = {10.1142/9789811218842_0007},
pages = {159--173},
}
@inproceedings{camps-valls_advancing_2020,
title = {{ADVANCING} {DEEP} {LEARNING} for {EARTH} {SCIENCES}: {From} {HYBRID} {MODELING} to {INTERPRETABILITY}},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85101964131&doi=10.1109%2fIGARSS39084.2020.9323558&partnerID=40&md5=29f8f67c094d60d61506e1b604c0d257},
doi = {10.1109/IGARSS39084.2020.9323558},
author = {Camps-Valls, G. and Reichstein, M. and Zhu, X. and Tuia, D.},
year = {2020},
pages = {3979--3982},
file = {Camps-Valls et al. - 2020 - ADVANCING DEEP LEARNING for EARTH SCIENCES From H.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\4PNZV7ZW\\Camps-Valls et al. - 2020 - ADVANCING DEEP LEARNING for EARTH SCIENCES From H.pdf:application/pdf},
}
@article{wei_meta-heuristic_2014,
title = {Meta-heuristic {Bayesian} networks retrieval combined polarization corrected temperature and scattering index for precipitations},
volume = {136},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84897962919&doi=10.1016%2fj.neucom.2014.01.030&partnerID=40&md5=89376d155b1b1a20cc4069947045332b},
doi = {10.1016/j.neucom.2014.01.030},
journal = {Neurocomputing},
author = {Wei, C.-C.},
year = {2014},
pages = {71--81},
file = {Wei - 2014 - Meta-heuristic Bayesian networks retrieval combine.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\WHEKMM2X\\Wei - 2014 - Meta-heuristic Bayesian networks retrieval combine.pdf:application/pdf},
}
@article{tesch_causal_2023,
title = {Causal deep learning models for studying the {Earth} system},
volume = {16},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85158145607&doi=10.5194%2fgmd-16-2149-2023&partnerID=40&md5=8ac89f5375d93dcd8928a8d05aa1d74e},
doi = {10.5194/gmd-16-2149-2023},
number = {8},
journal = {Geoscientific Model Development},
author = {Tesch, T. and Kollet, S. and Garcke, J.},
year = {2023},
pages = {2149--2166},
file = {Tesch et al. - 2023 - Causal deep learning models for studying the Earth.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\YMAMHUNY\\Tesch et al. - 2023 - Causal deep learning models for studying the Earth.pdf:application/pdf},
}
@article{luo_big_2019,
title = {Big data pioneers new ways of geoscience research: identifying relevant relationships to enhance research feasibility},
volume = {26},
shorttitle = {大数据开创地学研究新途径:查明相关关系,增强研究可行性},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85066411097&doi=10.13745%2fj.esf.sf.2019.4.28&partnerID=40&md5=10a38c83ad046c2bb70cef81a188794b},
doi = {10.13745/j.esf.sf.2019.4.28},
abstract = {Humans have entered the era of big data. Research ideas and methods based on big data have gained much attention and start to apply widely in the field of geoscience. In our view, the subject of big data research is data, the tool is the computer, the method and means are to find out the correlation between data, and the characteristics is to make decisions based on probability criteria. To reiterate: big data is the idea and method of finding out the correlation between data; it researches problems and make correct decisions by mining large amounts of data. In this paper, we suggest that the inductive method is the way to carry out big data research, specially as its research power has been greatly enhanced by high performance computer and big data technology. Through an in-depth analyses of statistics and machine learning algorithm, we came to the conclusion that big data shall change the ways people learn and understand nature and scientific studies are designed and performed. And it shall subvert the long-standing habit of conducting scientific research by finding causal relationships. Big data shall create a new approach to conducting geoscience research across complex causal relationships and obtaining research results directly. We concluded in this study that with the explosive growth of data, and with popularization of high-performance computers and rapid development of computing technology, the statistical analysis method has largely broken through the limitation of data volume. This shall enable statistical analysis and prediction models to generate truer thus more reliable results. Ultimately, the ability to explain conditions and outcomes, combining with the advantages of machine learning algorithms for semi-structured and unstructured data, will make quantitative geoscience research truly feasible.},
number = {4},
journal = {Earth Science Frontiers},
author = {Luo, J. and Zhang, Q.},
year = {2019},
pages = {6--12},
file = {Luo and Zhang - 2019 - Big data pioneers new ways of geoscience research.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\TAQXA637\\Luo and Zhang - 2019 - Big data pioneers new ways of geoscience research.pdf:application/pdf},
}
@inproceedings{jerzak_image-based_2023,
title = {Image-based {Treatment} {Effect} {Heterogeneity}},
volume = {213},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85147679914&partnerID=40&md5=7866b08ec210422443d7ad823941845f},
author = {Jerzak, C.T. and Johansson, F. and Daoud, A.},
year = {2023},
pages = {531--552},
file = {Jerzak et al. - 2023 - Image-based Treatment Effect Heterogeneity.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\VIN6FZMQ\\Jerzak et al. - 2023 - Image-based Treatment Effect Heterogeneity.pdf:application/pdf},
}
@article{yuan_understanding_2022,
title = {Understanding and reducing the uncertainties of land surface energy flux partitioning within {CMIP6} land models},
volume = {319},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85127533185&doi=10.1016%2fj.agrformet.2022.108920&partnerID=40&md5=4b8214c3e424650b8fff9e2f177da867},
doi = {10.1016/j.agrformet.2022.108920},
journal = {Agricultural and Forest Meteorology},
author = {Yuan, K. and Zhu, Q. and Riley, W.J. and Li, F. and Wu, H.},
year = {2022},
file = {Yuan et al. - 2022 - Understanding and reducing the uncertainties of la.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\7QIBC2Z6\\Yuan et al. - 2022 - Understanding and reducing the uncertainties of la.pdf:application/pdf},
}
@inproceedings{giannarakis_understanding_2023,
title = {{UNDERSTANDING} {THE} {IMPACTS} {OF} {CROP} {DIVERSIFICATION} {IN} {THE} {CONTEXT} {OF} {CLIMATE} {CHANGE}: {A} {MACHINE} {LEARNING} {APPROACH}},
volume = {48},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85183299841&doi=10.5194%2fisprs-archives-XLVIII-1-W2-2023-1379-2023&partnerID=40&md5=1ccdc9262ddae35138d24b0a67e73ed5},
doi = {10.5194/isprs-archives-XLVIII-1-W2-2023-1379-2023},
author = {Giannarakis, G. and Tsoumas, I. and Neophytides, S. and Papoutsa, C. and Kontoes, C. and Hadjimitsis, D.},
year = {2023},
note = {Issue: 1/W2-2023},
pages = {1379--1384},
file = {Giannarakis et al. - 2023 - UNDERSTANDING THE IMPACTS OF CROP DIVERSIFICATION .pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\4IIKIHIV\\Giannarakis et al. - 2023 - UNDERSTANDING THE IMPACTS OF CROP DIVERSIFICATION .pdf:application/pdf},
}
@article{sarkodie_causal_2020,
title = {Causal effect of environmental factors, economic indicators and domestic material consumption using frequency domain causality test},
volume = {736},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85085655127&doi=10.1016%2fj.scitotenv.2020.139602&partnerID=40&md5=12d752fde9b73d1ebcf3bcdb0d6be487},
doi = {10.1016/j.scitotenv.2020.139602},
journal = {Science of the Total Environment},
author = {Sarkodie, S.A.},
year = {2020},
file = {Sarkodie - 2020 - Causal effect of environmental factors, economic i.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\YTVKGUC2\\Sarkodie - 2020 - Causal effect of environmental factors, economic i.pdf:application/pdf},
}
@article{su_insight_2023,
title = {The insight of why: {Causal} inference in {Earth} system science},
volume = {66},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85169914547&doi=10.1007%2fs11430-023-1148-7&partnerID=40&md5=cbd25a4fdd6687041bad60d5f4dda225},
doi = {10.1007/s11430-023-1148-7},
number = {10},
journal = {Science China Earth Sciences},
author = {Su, J. and Chen, D. and Zheng, D. and Su, Y. and Li, X.},
year = {2023},
pages = {2169--2186},
file = {Su et al. - 2023 - The insight of why Causal inference in Earth syst.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\3P2XQ2UP\\Su et al. - 2023 - The insight of why Causal inference in Earth syst.pdf:application/pdf},
}
@article{li_big_2023,
title = {Big {Data} in {Earth} system science and progress towards a digital twin},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85156153784&doi=10.1038%2fs43017-023-00409-w&partnerID=40&md5=ae1ee1a5065cfd09c919f82fb008244a},
doi = {10.1038/s43017-023-00409-w},
journal = {Nature Reviews Earth and Environment},
author = {Li, X. and Feng, M. and Ran, Y. and Su, Y. and Liu, F. and Huang, C. and Shen, H. and Xiao, Q. and Su, J. and Yuan, S. and Guo, H.},
year = {2023},
file = {Li et al. - 2023 - Big Data in Earth system science and progress towa.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\6FFYHVUA\\Li et al. - 2023 - Big Data in Earth system science and progress towa.pdf:application/pdf},
}
@article{li_causal_2020,
title = {A causal inference model based on random forests to identify the effect of soil moisture on precipitation},
volume = {21},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85085945542&doi=10.1175%2fJHM-D-19-0209.1&partnerID=40&md5=400103a6dfcb6927cbb4842bf3f6e2de},
doi = {10.1175/JHM-D-19-0209.1},
number = {5},
journal = {Journal of Hydrometeorology},
author = {Li, L.U. and Shangguan, W.E.I. and Deng, Y.I. and Mao, J. and Pan, J. and Wei, N.A.N. and Yuan, H.U.A. and Zhang, S. and Zhang, Y. and Dai, Y.},
year = {2020},
pages = {1115--1131},
file = {Li et al. - 2020 - A causal inference model based on random forests t.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\B34V2UW9\\Li et al. - 2020 - A causal inference model based on random forests t.pdf:application/pdf},
}
@article{harding_small_2021,
title = {Small steps with big data: {Using} machine learning in energy and environmental economics},
volume = {13},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85117145366&doi=10.1146%2fannurev-resource-100920-034117&partnerID=40&md5=c51dad35663dee389acfa915bca1971c},
doi = {10.1146/annurev-resource-100920-034117},
journal = {Annual Review of Resource Economics},
author = {Harding, M.C. and Lamarche, C.},
year = {2021},
pages = {469--488},
file = {Harding and Lamarche - 2021 - Small steps with big data Using machine learning .pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\GGYG3GBS\\Harding and Lamarche - 2021 - Small steps with big data Using machine learning .pdf:application/pdf},
}
@inproceedings{runge_causality_2019,
title = {The {Causality} for {Climate} {Competition}},
volume = {123},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85102541381&partnerID=40&md5=12adb7bf6e4901f53374a3d818bf7f19},
author = {Runge, J. and Tibau, X.-A. and Bruhns, M. and Muñoz-Marí, J. and Camps-Valls, G.},
year = {2019},
pages = {110--120},
file = {Runge et al. - 2019 - The Causality for Climate Competition.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\KCIK76CT\\Runge et al. - 2019 - The Causality for Climate Competition.pdf:application/pdf},
}
@article{tan_correlation_2011,
title = {Correlation of {Asian} dust with chlorophyll and primary productivity in the coastal seas of {China} during the period from 1998 to 2008},
volume = {116},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-79960190981&doi=10.1029%2f2010JG001456&partnerID=40&md5=ada44997def422ec1f1c5081a93c0a3d},
doi = {10.1029/2010JG001456},
number = {2},
journal = {Journal of Geophysical Research: Biogeosciences},
author = {Tan, S.-C. and Shi, G.-Y. and Shi, J.-H. and Gao, H.-W. and Yao, X.},
year = {2011},
file = {Tan et al. - 2011 - Correlation of Asian dust with chlorophyll and pri.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\ZI94FJDU\\Tan et al. - 2011 - Correlation of Asian dust with chlorophyll and pri.pdf:application/pdf},
}
@article{noy_long-term_2021,
title = {A long-term spatiotemporal analysis of biocrusts across a diverse arid environment: {The} case of the {Israeli}-{Egyptian} sandfield},
volume = {774},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85101340613&doi=10.1016%2fj.scitotenv.2021.145154&partnerID=40&md5=1362384f34f2d55fe3f3157ff7165bba},
doi = {10.1016/j.scitotenv.2021.145154},
journal = {Science of the Total Environment},
author = {Noy, K. and Ohana-Levi, N. and Panov, N. and Silver, M. and Karnieli, A.},
year = {2021},
keywords = {Remote sensing, BIOLOGICAL SOIL CRUSTS, CLUSTERING METHOD, COVER TRANSITIONS, Crust Index, LAND-SURFACE TEMPERATURE, Landsat, Long-term trend, NEGEV DESERT, RAINFALL GRADIENT, SAND, SPECTRAL REFLECTANCE, TIME-SERIES ANALYSIS, Time-series clustering, VEGETATION-COVER},
file = {Noy et al. - 2021 - A long-term spatiotemporal analysis of biocrusts a.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\J36ZRC8V\\Noy et al. - 2021 - A long-term spatiotemporal analysis of biocrusts a.pdf:application/pdf},
}
@article{das_deep-learning-based_2017,
title = {A {Deep}-{Learning}-{Based} {Forecasting} {Ensemble} to {Predict} {Missing} {Data} for {Remote} {Sensing} {Analysis}},
volume = {10},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85039446572&doi=10.1109%2fJSTARS.2017.2760202&partnerID=40&md5=8fe5d528459684d10cb1a2187e3eef63},
doi = {10.1109/JSTARS.2017.2760202},
number = {12},
journal = {IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing},
author = {Das, M. and Ghosh, S.K.},
year = {2017},
keywords = {deep learning, Machine learning, Predictive models, remote sensing, Causality constraint, ensemble, Image reconstruction, missing data, prediction, Spatiotemporal phenomena, TIme series analysis},
pages = {5228--5236},
file = {Das and Ghosh - 2017 - A Deep-Learning-Based Forecasting Ensemble to Pred.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\CYJADNLR\\Das and Ghosh - 2017 - A Deep-Learning-Based Forecasting Ensemble to Pred.pdf:application/pdf},
}
@article{zheng_using_2016,
title = {Using machine learning in environmental tax reform assessment for sustainable development: {A} case study of {Hubei} {Province}, {China}},
volume = {8},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85000715918&doi=10.3390%2fsu8111124&partnerID=40&md5=c7ba4f582d938e997e02930050259603},
doi = {10.3390/su8111124},
number = {11},
journal = {Sustainability (Switzerland)},
author = {Zheng, Y. and Zheng, H. and Ye, X.},
year = {2016},
file = {Zheng et al. - 2016 - Using machine learning in environmental tax reform.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\PD58DNR3\\Zheng et al. - 2016 - Using machine learning in environmental tax reform.pdf:application/pdf},
}
@article{ji_river_2024,
title = {River runoff causal discovery with deep reinforcement learning},
volume = {54},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85186573667&doi=10.1007%2fs10489-024-05348-7&partnerID=40&md5=389874488fc3ed6e2b3640283e40af0d},
doi = {10.1007/s10489-024-05348-7},
number = {4},
journal = {Applied Intelligence},
author = {Ji, J. and Wang, T. and Liu, J. and Wang, M. and Tang, W.},
year = {2024},
pages = {3547--3565},
file = {Ji et al. - 2024 - River runoff causal discovery with deep reinforcem.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\YPMYM3CS\\Ji et al. - 2024 - River runoff causal discovery with deep reinforcem.pdf:application/pdf},
}
@inproceedings{ali_quantifying_2023,
title = {Quantifying {Causes} of {Arctic} {Amplification} via {Deep} {Learning} {Based} {Time}-{Series} {Causal} {Inference}},
url = {https://ieeexplore.ieee.org/document/10460053},
doi = {10.1109/ICMLA58977.2023.00101},
abstract = {The warming of the Arctic, also known as Arctic amplification, is led by several atmospheric and oceanic drivers. However, the details of its underlying thermodynamic causes are still unknown. Inferring the causal effects of atmospheric processes on sea ice melt using fixed treatment effect strategies leads to unrealistic counterfactual estimations. Such methods are also prone to bias due to time-varying confoundedness. Further, the complex non-linearity in Earth science data makes it infeasible to perform causal inference using existing marginal structural techniques. In order to tackle these challenges, we propose TCINet - Time-series Causal Inference Network to infer causation under continuous treatment using recurrent neural networks and a novel probabilistic balancing technique. More specifically, we propose a neural network based potential outcome model using the long-short-term-memory (LSTM) layers for time-delayed factual and counterfactual predictions with a custom weighted loss. To tackle the confounding bias, we experiment with multiple balancing strategies, namely TCINet with the inverse probability weighting (IPTW), TCINet with stabilized weights using Gaussian Mixture Model (GMMs) and TCINet without any balancing technique. Through experiments on synthetic and observational data, we show how our research can substantially improve the ability to quantify leading causes of Arctic sea ice melt, further paving paths for causal inference in observational Earth science.},
urldate = {2024-04-16},
booktitle = {2023 {International} {Conference} on {Machine} {Learning} and {Applications} ({ICMLA})},
author = {Ali, Sahara and Faruque, Omar and Huang, Yiyi and Gani, Md. Osman and Subramanian, Aneesh and Schlegel, Nicole-Jeanne and Wang, Jianwu},
month = dec,
year = {2023},
note = {ISSN: 1946-0759},
keywords = {Deep Learning, Deep learning, Arctic, Arctic Amplification, Causal Inference, Estimation, LSTM, Predictive models, Probabilistic logic, Recurrent neural networks, Thermodynamics},
pages = {689--696},
file = {IEEE Xplore Full Text PDF:C\:\\Users\\kazsa11\\Zotero\\storage\\5HYTZSEA\\Ali et al. - 2023 - Quantifying Causes of Arctic Amplification via Dee.pdf:application/pdf},
}
@article{dong_novel_2024,
title = {A {Novel} {Causal} {Inference}-{Guided} {Feature} {Enhancement} {Framework} for {PolSAR} {Image} {Classification}},
volume = {62},
issn = {1558-0644},
url = {https://ieeexplore.ieee.org/document/10360854},
doi = {10.1109/TGRS.2023.3343380},
abstract = {In recent years, there has been a prominent focus on enhancing the quality of features derived from convolutional neural networks (CNNs) within the field of polarimetric synthetic aperture radar (PolSAR) image classification. Targeting this challenge, this article first visualizes the lack of discriminability and generalizability in CNN features through several empirical observations. Subsequently, we explain why these problems arise from a causal perspective, accomplished by means of a structural causal model (SCM) constructed according to the training and testing process of CNNs. This SCM facilitates the identification of variables that affect the quality of PolSAR image feature learning, as well as an intervention on those variables using backdoor adjustment. Building upon this groundwork, a novel causal inference-guided feature enhancement framework is constructed. It can be seamlessly integrated into any CNN-based PolSAR image classifier in a plug-and-play manner, enabling the enhanced classifier to filter out interference information and prevent model overfitting. These two aspects bring better feature discriminability and generalizability, respectively, leading to improved classification performance. Experimental results on four widely-used PolSAR image datasets demonstrate the effectiveness of our proposed framework. We integrate it into several mainstream methods in the field and show that the accuracy of the enhanced classifier is improved compared to the original model.},
urldate = {2024-04-16},
journal = {IEEE Transactions on Geoscience and Remote Sensing},
author = {Dong, Hongwei and Si, Lingyu and Qiang, Wenwen and Zhang, Lamei and Yu, Junzhi and Wu, Yuquan and Zheng, Changwen and Sun, Fuchun},
year = {2024},
note = {Conference Name: IEEE Transactions on Geoscience and Remote Sensing},
keywords = {deep learning, Remote sensing, Causal inference, Convolution, Convolutional neural networks, feature enhancement, Feature extraction, image classification, Image classification, polarimetric synthetic aperture radar (PolSAR), Synthetic aperture radar, Training},
pages = {1--16},
file = {IEEE Xplore Full Text PDF:C\:\\Users\\kazsa11\\Zotero\\storage\\YPIDMNHE\\Dong et al. - 2024 - A Novel Causal Inference-Guided Feature Enhancemen.pdf:application/pdf},
}
@article{wang_visual_2016,
title = {The {Visual} {Causality} {Analyst}: {An} {Interactive} {Interface} for {Causal} {Reasoning}},
volume = {22},
issn = {1941-0506},
shorttitle = {The {Visual} {Causality} {Analyst}},
url = {https://ieeexplore.ieee.org/document/7192729},
doi = {10.1109/TVCG.2015.2467931},
abstract = {Uncovering the causal relations that exist among variables in multivariate datasets is one of the ultimate goals in data analytics. Causation is related to correlation but correlation does not imply causation. While a number of casual discovery algorithms have been devised that eliminate spurious correlations from a network, there are no guarantees that all of the inferred causations are indeed true. Hence, bringing a domain expert into the casual reasoning loop can be of great benefit in identifying erroneous casual relationships suggested by the discovery algorithm. To address this need we present the Visual Causal Analyst - a novel visual causal reasoning framework that allows users to apply their expertise, verify and edit causal links, and collaborate with the causal discovery algorithm to identify a valid causal network. Its interface consists of both an interactive 2D graph view and a numerical presentation of salient statistical parameters, such as regression coefficients, p-values, and others. Both help users in gaining a good understanding of the landscape of causal structures particularly when the number of variables is large. Our framework is also novel in that it can handle both numerical and categorical variables within one unified model and return plausible results. We demonstrate its use via a set of case studies using multiple practical datasets.},
number = {1},
urldate = {2024-04-16},
journal = {IEEE Transactions on Visualization and Computer Graphics},
author = {Wang, Jun and Mueller, Klaus},
month = jan,
year = {2016},
note = {Conference Name: IEEE Transactions on Visualization and Computer Graphics},
keywords = {Causality, Correlation, High-dimensional data, Hypothesis testing, Inference algorithms, Layout, Linear regression, Optimization, Visual evidence, Visual knowledge discovery, Visualization},
pages = {230--239},
file = {IEEE Xplore Full Text PDF:C\:\\Users\\kazsa11\\Zotero\\storage\\T72NMGIA\\Wang and Mueller - 2016 - The Visual Causality Analyst An Interactive Inter.pdf:application/pdf},
}
@article{tuia_toward_2021,
title = {Toward a {Collective} {Agenda} on {AI} for {Earth} {Science} {Data} {Analysis}},
volume = {9},
issn = {2168-6831},
url = {https://ieeexplore.ieee.org/document/9456877},
doi = {10.1109/MGRS.2020.3043504},
abstract = {In past years, we have witnessed the fields of geosciences and remote sensing and artificial intelligence (AI) become closer. Thanks to the massive availability of observational data, improved simulations, and algorithmic advances, these disciplines have found common objectives and challenges to help advance the modeling and understanding of the Earth system. Despite such great opportunities, we have also observed a worrisome tendency to remain in disciplinary comfort zones, applying recent advances from AI on well-resolved remote sensing problems. Here, we take a position on the research directions for which we think the interface between these fields will have the most significant impact and become potential game changers. In our declared agenda for AI in Earth sciences, we aim to inspire researchers, especially the younger generations, to tackle these challenges for a real advance of remote sensing and the geosciences.},
number = {2},
urldate = {2024-04-16},
journal = {IEEE Geoscience and Remote Sensing Magazine},
author = {Tuia, Devis and Roscher, Ribana and Wegner, Jan Dirk and Jacobs, Nathan and Zhu, Xiaoxiang and Camps-Valls, Gustau},
month = jun,
year = {2021},
note = {Conference Name: IEEE Geoscience and Remote Sensing Magazine},
keywords = {Artificial intelligence, Data analysis, Data models, Earth science, Games, Geology},
pages = {88--104},
file = {IEEE Xplore Full Text PDF:C\:\\Users\\kazsa11\\Zotero\\storage\\EQR9XYX9\\Tuia et al. - 2021 - Toward a Collective Agenda on AI for Earth Science.pdf:application/pdf},
}
@article{perez-suay_causal_2019,
title = {Causal {Inference} in {Geoscience} and {Remote} {Sensing} {From} {Observational} {Data}},
volume = {57},
issn = {1558-0644},
url = {https://ieeexplore.ieee.org/document/8475013},
doi = {10.1109/TGRS.2018.2867002},
abstract = {Establishing causal relations between random variables from observational data is perhaps the most important challenge in today's science. In remote sensing and geosciences, this is of special relevance to better understand the earth's system and the complex interactions between the governing processes. In this paper, we focus on an observational causal inference, and thus, we try to estimate the correct direction of causation using a finite set of empirical data. In addition, we focus on the more complex bivariate scenario that requires strong assumptions and no conditional independence tests can be used. In particular, we explore the framework of (nondeterministic) additive noise models, which relies on the principle of independence between the cause and the generating mechanism. A practical algorithmic instantiation of such principle only requires: 1) two regression models in the forward and backward directions and 2) the estimation of statistical independence between the obtained residuals and the observations. The direction leading to more independent residuals is decided to be the cause. We instead propose a criterion that uses the sensitivity (derivative) of the dependence estimator, the sensitivity criterion allows to identify samples most affecting the dependence measure, and hence, the criterion is robust to spurious detections. We illustrate the performance in a collection of 28 geoscience causal inference problems, a database of radiative transfer models simulations and machine learning emulators in vegetation parameter modeling involving 182 problems, and assessing the impact of different regression models in a carbon cycle problem. The criterion achieves the state-of-the-art detection rates in all cases, and it is generally robust to noise sources and distortions. The presented approach confirms the validity in observational bivariate problems in the earth sciences.},
number = {3},
urldate = {2024-04-16},
journal = {IEEE Transactions on Geoscience and Remote Sensing},
author = {Pérez-Suay, Adrián and Camps-Valls, Gustau},
month = mar,
year = {2019},
note = {Conference Name: IEEE Transactions on Geoscience and Remote Sensing},
keywords = {Remote sensing, Causal inference, Geology, Biological system modeling, dependence estimation, Earth, Gaussian process (GP) regression, Hilbert–Schmidt independence criterion (HSIC), Mathematical model, Meteorology, noise, sensitivity, Sensitivity},
pages = {1502--1513},
file = {IEEE Xplore Full Text PDF:C\:\\Users\\kazsa11\\Zotero\\storage\\K62Z5AAQ\\Pérez-Suay and Camps-Valls - 2019 - Causal Inference in Geoscience and Remote Sensing .pdf:application/pdf},
}
@inproceedings{otgonbaatar_causality_2022,
title = {Causality for {Remote} {Sensing}: {An} {Exploratory} {Study}},
shorttitle = {Causality for {Remote} {Sensing}},
url = {https://ieeexplore.ieee.org/document/9883060},
doi = {10.1109/IGARSS46834.2022.9883060},
abstract = {Causality is one of the most important topics in a Machine Learning (ML) research, and it gives insights beyond the dependency of data points. Causality is a very vital concept also for investigating the dynamic surface of our living planet. However, there are not many attempts for integrating a causal model in Remote Sensing (RS) methodologies. Hence, in this paper, we propose to use patch-based RS images and to represent each patch-based image by a single variable (e.g. entropy). Then we use a Structural Equation Model (SEM) to study their cause-effect relation. Moreover, the SEM is a simple causal model characterized by a Directed Acyclic Graph (DAG). Its nodes are causal variables, and its edges represent causal relationships among causal variables if and only if causal variables are dependent.},
urldate = {2024-04-16},
booktitle = {{IGARSS} 2022 - 2022 {IEEE} {International} {Geoscience} and {Remote} {Sensing} {Symposium}},
author = {Otgonbaatar, Soronzonbold and Datcu, Mihai and Demir, Begüm},
month = jul,
year = {2022},
note = {ISSN: 2153-7003},
keywords = {machine learning, Agriculture, causality, earth observation, Image edge detection, Machine learning algorithms, Mathematical models, Numerical analysis, Planets, remote sensing, Toy manufacturing industry},
pages = {259--262},
file = {IEEE Xplore Full Text PDF:C\:\\Users\\kazsa11\\Zotero\\storage\\9BXCT7IG\\Otgonbaatar et al. - 2022 - Causality for Remote Sensing An Exploratory Study.pdf:application/pdf},
}
@inproceedings{qiu_economic_2021,
title = {Economic {Development} {Analysis} of the {Belt} and {Road} {Regions} {Based} on {Automatic} {Interpretation} of {Remote} {Sensing} {Images}},
url = {https://ieeexplore.ieee.org/document/9660561},
doi = {10.1109/IC-NIDC54101.2021.9660561},
abstract = {The Belt and Road (B\&R) initiative is proposed to promote common development among countries along the B\&R. In recent years, although the B\&R has contributed to the regions along the route, it is always a controversial topic in the international community. A number of scholars have done a set of research works to analyze the effects of the B\&R projects based on traditional economic methods. However, the drawbacks of subjectivity and delay reduce the conviction of the analysis results. In this paper, we leverage the objectivity and real-time features of remote sensing (RS) images to analyze the effects of the B\&R project. Our research takes Voi town along the Mongolia-Nairobi Railway as the representative city. In addition, in order to prove the causal relationship between the B\&r; economic development, we select the Taveta town as the comparison city. The semantic segmentation based on deep learning is applied to the multi-temporal RS images, to retrieve the economic development by automatically recognizing houses. On this basis, the construction and development of both the studied region and the comparison are quantitatively analyzed by meshing analysis and standard deviation elliptic methods. For overcoming the shortages of the conventional algorithms, a novel segmentation network based on the attention mechanism is proposed. The evaluation proves the semantic segmentation results can fully support the follow-up data analysis. In addition, the analysis results show that our work is a convincing initiative to reveal the values of the B\&R projects for economic developments in the B\&R-related regions.},
urldate = {2024-04-16},
booktitle = {2021 7th {IEEE} {International} {Conference} on {Network} {Intelligence} and {Digital} {Content} ({IC}-{NIDC})},
author = {Qiu, Xinzhu and Wang, Yunzhe and Cao, Jingyi and Xu, Guannan and You, Yanan and Ren, Junlong},
month = nov,
year = {2021},
note = {ISSN: 2575-4955},
keywords = {Semantics, Roads, Urban areas, Artificial Intelligence, Belts, Economic Development Analysis, Economics, Image segmentation, Remote Sensing, Semantic Segmentation, Statistical analysis, the Belt and Road},
pages = {96--101},
file = {Qiu et al. - 2021 - Economic Development Analysis of the Belt and Road.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\5W2N48U6\\Qiu et al. - 2021 - Economic Development Analysis of the Belt and Road.pdf:application/pdf},
}
@inproceedings{li_urban_2019,
title = {Urban land price assessment based on {GIS} and deep learning},
url = {https://ieeexplore.ieee.org/document/8900516},
doi = {10.1109/IGARSS.2019.8900516},
abstract = {The land price reflects the supply and demand of the land market and the economic life of the city, is indispensable to regulate urban land use and optimize the allocation of land resources. Due to the complex factors affecting the price of urban land, involving natural factors, social factors, economic factors, market factors, there is currently no model method at home and abroad that can effectively integrate these factors for residential land price assessment.This research explore the identification method of urban land price influencing factors in artificial intelligence environment, and combine deep learning algorithm with urban land price evaluation method. The deep neural network is used to integrate the spatial characteristics of land influencing factors. By establishing the deep hybrid neural network with space features, the linear relationship and causal relationship of influencing factors and the land price are automatically identified. The deep learning algorithm for the factors affecting of Shenzhen urban land price, promote the intelligent evaluation of land price.},
urldate = {2024-04-16},
booktitle = {{IGARSS} 2019 - 2019 {IEEE} {International} {Geoscience} and {Remote} {Sensing} {Symposium}},
author = {Li, Hongga and Huang, Xiaoxia and Li, Xia},
month = jul,
year = {2019},
note = {ISSN: 2153-7003},
keywords = {Deep learning, deep learning, Training, Analytical models, Urban areas, Economics, Biological neural networks, GIS, land price assessment},
pages = {935--938},
file = {Li et al. - 2019 - Urban land price assessment based on GIS and deep .pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\XDDXECT3\\Li et al. - 2019 - Urban land price assessment based on GIS and deep .pdf:application/pdf},
}
@article{xiong_explainable_2022,
title = {An {Explainable} {Attention} {Network} for {Fine}-{Grained} {Ship} {Classification} {Using} {Remote}-{Sensing} {Images}},
volume = {60},
issn = {1558-0644},
url = {https://ieeexplore.ieee.org/document/9741720},
doi = {10.1109/TGRS.2022.3162195},
abstract = {Advances in space-based ocean surveillance systems have improved the detection of objects from high-quality remote-sensing big data. Previous studies mainly focused on finding and recognizing objects based on deep learning and statistical frameworks. Studies have not fully explored the transparent and reasonable decision-making process for the final predicted results, which is vital for civil and military applications. An explainable attention network for fine-grained ship image classification is proposed in the present study to bridge this gap. The present study seeks to increase attention to objects’ discriminative parts and explore intrinsic relationships between multiple attention parts and predicted outcomes. Several causal multi-attention maps are generated by combining the multi-head attention mechanism and structural causal model. The convolutional filters in the last layer of the network are divided into several groups, and each group is designed to express specific semantic information under supervision of the filter loss function. The results show which parts of the objects are adopted as the key factors for the network to make the final predicted outcome. In the training process, the network is designed to rapidly focus on the salient feature of objects and played a role in guiding other parts of the network to improve the explainable capability of the network without affecting the discrimination power or compromising the classification accuracy. Extensive experiments based on two public datasets show that the network is highly effective as indicated by high classification accuracy and explainable ability.},
urldate = {2024-04-16},
journal = {IEEE Transactions on Geoscience and Remote Sensing},
author = {Xiong, Wei and Xiong, Zhenyu and Cui, Yaqi},
year = {2022},
note = {Conference Name: IEEE Transactions on Geoscience and Remote Sensing},
keywords = {Remote sensing, Predictive models, Causal inference, Feature extraction, Training, Visualization, remote sensing, Task analysis, explainable visual attention, fine-grained ship classification, Marine vehicles},
pages = {1--14},
file = {Xiong et al. - 2022 - An Explainable Attention Network for Fine-Grained .pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\3BP3AUJV\\Xiong et al. - 2022 - An Explainable Attention Network for Fine-Grained .pdf:application/pdf},
}
@inproceedings{wang_remote_2019,
title = {Remote {Sensing} {Image} {Synthesis} via {Graphical} {Generative} {Adversarial} {Networks}},
url = {https://ieeexplore.ieee.org/document/8898915},
doi = {10.1109/IGARSS.2019.8898915},
abstract = {We explore the use of graphical generative adversarial networks (Graphical-GAN) for synthesizing remote sensing images. The model is probabilistic graphical based generative adversarial networks (GAN). It pairs a generative network G with a recognition network R. Both of them are adversarially trained with a discriminative network D. Particularly, R is employed to infer the underlying causal relationships among both observed and latent variables from real remote sensing images. The advantages of the Graphical-GAN for synthesizing multiple categories of remote sensing images are two fold. Firstly, it considers the underlying causal relationships and captures the true data distribution of remote sensing images. Secondly, the adversarial learning generates synthetic sensing images that are similar to real ones with slight differences. Our remote sensing image synthesis scheme paves a promising way for remote sensing dataset augmentation, which is an effective means of improving the accuracy of learning models. Experimental results with high Inception Scores (IS) validate the effectiveness of the Graphical-GAN for remote sensing image synthesis.},
urldate = {2024-04-16},
booktitle = {{IGARSS} 2019 - 2019 {IEEE} {International} {Geoscience} and {Remote} {Sensing} {Symposium}},
author = {Wang, Guangxing and Dong, Guoshuai and Li, Hui and Han, Lirong and Tao, Xuanwen and Ren, Peng},
month = jul,
year = {2019},
note = {ISSN: 2153-7003},
keywords = {Remote sensing, Probabilistic logic, Training, Bayes methods, Generative adversarial networks, Graphical Generative Adversarial Networks, Image synthesis, Neural networks, Remote Sensing Image Synthesis},
pages = {10027--10030},
file = {Wang et al. - 2019 - Remote Sensing Image Synthesis via Graphical Gener.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\6TLALUYJ\\Wang et al. - 2019 - Remote Sensing Image Synthesis via Graphical Gener.pdf:application/pdf},
}
@inproceedings{giannarakis_towards_2022,
title = {Towards assessing agricultural land suitability with causal machine learning},
url = {https://ieeexplore.ieee.org/document/9856946},
doi = {10.1109/CVPRW56347.2022.00150},
abstract = {Understanding the suitability of agricultural land for applying specific management practices is of great importance for sustainable and resilient agriculture against climate change. Recent developments in the field of causal machine learning enable the estimation of intervention impacts on an outcome of interest, for samples described by a set of observed characteristics. We introduce an extensible data-driven framework that leverages earth observations and frames agricultural land suitability as a geospatial impact assessment problem, where the estimated effects of agricultural practices on agroecosystems serve as a land suitability score and guide decision making. We formulate this as a causal machine learning task and discuss how this approach can be used for agricultural planning in a changing climate. Specifically, we extract the agricultural management practices of "crop rotation" and "landscape crop diversity" from crop type maps, account for climate and land use data, and use double machine learning to estimate their heterogeneous effect on Net Primary Productivity (NPP), within the Flanders region of Belgium from 2010 to 2020. We find that the effect of crop rotation was insignificant, while landscape crop diversity had a small negative effect on NPP. Finally, we observe considerable effect heterogeneity in space for both practices and analyze it.},
urldate = {2024-04-16},
booktitle = {2022 {IEEE}/{CVF} {Conference} on {Computer} {Vision} and {Pattern} {Recognition} {Workshops} ({CVPRW})},
author = {Giannarakis, Georgios and Sitokonstantinou, Vasileios and Lorilla, Roxanne Suzette and Kontoes, Charalampos},
month = jun,
year = {2022},
note = {ISSN: 2160-7516},
keywords = {Machine learning, Earth, Climate change, Crops, Decision making, Liquid crystal displays, Regulation},
pages = {1441--1451},
file = {Submitted Version:C\:\\Users\\kazsa11\\Zotero\\storage\\5WBRD2JT\\Giannarakis et al. - 2022 - Towards assessing agricultural land suitability wi.pdf:application/pdf},
}
@article{rong_information_2022,
title = {An {Information} {Flow}-{Based} {Sea} {Surface} {Height} {Reconstruction} {Through} {Machine} {Learning}},
volume = {60},
issn = {1558-0644},
url = {https://ieeexplore.ieee.org/document/9669264},
doi = {10.1109/TGRS.2022.3140398},
abstract = {The advent of satellite altimetry datasets of sea surface height (SSH) is a major advance in oceanography and other Earth system sciences. However, while the along-track data coverage is dense, the relatively poor resolution between tracks poses a challenge to the reconstruction of those processes such as mesoscale and submesoscale eddies. This study proposes a machine learning algorithm based on a causal inference tool, i.e., the Liang–Kleeman information flow (L-K IF) analysis, to address the challenge. For a region in the South China Sea where eddies frequently appear but unobserved, it is shown that the algorithm can reconstruct the desired mesoscale eddies in a remarkably successful way in geometry, orientation, strength, etc., while with the objective analysis interpolation or the traditional neural network technique, the results are not satisfactory. This study provides prospects for developing the next generation of SSH products with the available altimetry data.},
urldate = {2024-04-16},
journal = {IEEE Transactions on Geoscience and Remote Sensing},
author = {Rong, Yineng and Liang, X. San},
year = {2022},
note = {Conference Name: IEEE Transactions on Geoscience and Remote Sensing},
keywords = {machine learning, Causal inference, Data models, Satellites, Artificial neural networks, Interpolation, Liang–Kleeman information flow (L-K IF), Sea surface, sea surface height (SSH), Spatial resolution, Surface reconstruction},
pages = {1--9},
file = {Rong and Liang - 2022 - An Information Flow-Based Sea Surface Height Recon.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\KJY846VW\\Rong and Liang - 2022 - An Information Flow-Based Sea Surface Height Recon.pdf:application/pdf},
}
@article{xu_texture-aware_2024,
title = {Texture-{Aware} {Causal} {Feature} {Extraction} {Network} for {Multimodal} {Remote} {Sensing} {Data} {Classification}},
volume = {62},
issn = {1558-0644},
url = {https://ieeexplore.ieee.org/document/10440611},
doi = {10.1109/TGRS.2024.3368091},
abstract = {The pixel-level classification of multimodal remote sensing (RS) images plays a crucial role in the intelligent interpretation of RS data. However, existing methods that mainly focus on feature interaction and fusion fail to address the challenges posed by confounders—brought by sensor imaging bias, limiting their performance. In this article, we introduce causal inference into intelligent interpretation of RS and propose a new texture-aware causal feature extraction network (TeACFNet) for pixel-level fusion classification. Specifically, we propose a two-stage causal feature extraction (CFE) framework that helps networks learn more explicit class representations by capturing the causal relationships between multimodal heterogeneous data. In addition, to solve the problem of low-resolution land cover feature representation in RS images, we propose the refined statistical texture extraction (ReSTE) module. This module integrates the semantics of statistical textures in shallow feature maps through feature refinement, quantization, and encoding. Extensive experiments on two publicly available datasets with different modalities, including Houston2013 and Berlin datasets, demonstrate the remarkable effectiveness of our proposed method, which reaches a new state-of-the-art.},
urldate = {2024-04-16},
journal = {IEEE Transactions on Geoscience and Remote Sensing},
author = {Xu, Zhengyi and Jiang, Wen and Geng, Jie},
year = {2024},
note = {Conference Name: IEEE Transactions on Geoscience and Remote Sensing},
keywords = {Remote sensing, Feature extraction, Semantics, Causal feature extraction, Data mining, feature fusion, image pixel-level classification, Imaging, Land surface, multimodal remote sensing (RS), Representation learning, texture representation learning},
pages = {1--12},
file = {Xu et al. - 2024 - Texture-Aware Causal Feature Extraction Network fo.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\3HN834VB\\Xu et al. - 2024 - Texture-Aware Causal Feature Extraction Network fo.pdf:application/pdf},
}
@inproceedings{camps-valls_physics-aware_2021,
title = {Physics-{Aware} {Machine} {Learning} for {Geosciences} and {Remote} {Sensing}},
url = {https://ieeexplore.ieee.org/document/9554521},
doi = {10.1109/IGARSS47720.2021.9554521},
abstract = {Machine learning models alone are excellent approximators, but very often do not respect the most elementary laws of physics, like mass or energy conservation, so consistency and confidence are compromised. In this paper we describe the main challenges ahead in the field, and introduce several ways to live in the Physics and machine learning interplay: encoding differential equations from data, constraining data-driven models with physics-priors and dependence constraints, improving parameterizations, emulating physical models, and blending data-driven and process-based models. This is a collective long-term AI agenda towards developing and applying algorithms capable of discovering knowledge in the Earth system.},
urldate = {2024-04-16},
booktitle = {2021 {IEEE} {International} {Geoscience} and {Remote} {Sensing} {Symposium} {IGARSS}},
author = {Camps-Valls, Gustau and Svendsen, Daniel H. and Cortés-Andrés, Jordi and Mareno-Martínez, Álvaro and Pérez-Suay, Adrián and Adsuara, Jose and Martín, Irene and Piles, Maria and Muñoz-Marí, Jordi and Martino, Luca},
month = jul,
year = {2021},
note = {ISSN: 2153-7003},
keywords = {Deep learning, Machine learning, Data models, Geology, Earth, causality, Machine learning algorithms, Mathematical models, Energy conservation, hybrid machine learning, interpretability},
pages = {2086--2089},
file = {Camps-Valls et al. - 2021 - Physics-Aware Machine Learning for Geosciences and.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\J6V3EHA6\\Camps-Valls et al. - 2021 - Physics-Aware Machine Learning for Geosciences and.pdf:application/pdf},
}
@article{zhang_unsupervised_2022,
title = {Unsupervised {SAR} and {Optical} {Image} {Matching} {Using} {Siamese} {Domain} {Adaptation}},
volume = {60},
issn = {1558-0644},
url = {https://ieeexplore.ieee.org/document/9762950},
doi = {10.1109/TGRS.2022.3170316},
abstract = {Due to its highly complementary information about remote sensing, synthetic aperture radar (SAR) and optical imagery matching have drawn much attention in recent years. Compared with traditional methods, deep learning-based SAR-optical image matching models largely rely on supervision with ground truths, where the matching accuracy suffers because of unseen image domains. To mitigate loads in burdensome labeling tasks, transferring deep learning models trained with annotated source domains to nonannotated target domains has attracted great concern. Due to the domain gap, the difference between the source and target domains is likely to deteriorate the matching accuracy on target data if the training process is directly conducted without proper domain adaptation (DA). In this research, a Siamese DA (SDA) approach with a combined loss function is developed in the context of multimodality image matching. Then, a novel rotation/scale-invariant transformation module with regression modules is designed to extract rotation/scale-equivariant features. Finally, the causal inference-based self-learning method and the multiresolution histogram matching approach are employed to enhance the unsupervised matching performance. Experimental results on the RadarSat/Planet dataset and the Sentinel-1/2 dataset demonstrate that the developed model can achieve competitive matching performance with a low overlap ratio between domains and little data labeling. By alleviating the domain discrepancy, the developed model drastically reduces the average L2 score of the unsupervised matching from 9.576 to 0.658, while the less-than-one-pixel matching error rate is enhanced from 66.3\% to 90.6\%.},
urldate = {2024-04-16},
journal = {IEEE Transactions on Geoscience and Remote Sensing},
author = {Zhang, Zhaoxiang and Xu, Yuelei and Cui, Qi and Zhou, Qing and Ma, Linhua},
year = {2022},
note = {Conference Name: IEEE Transactions on Geoscience and Remote Sensing},
keywords = {Feature extraction, Synthetic aperture radar, synthetic aperture radar (SAR), Optical sensors, Adaptation models, Adaptive optics, Domain adaptation (DA), image matching, Image registration, Optical imaging, Siamese},
pages = {1--16},
file = {Zhang et al. - 2022 - Unsupervised SAR and Optical Image Matching Using .pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\VDAEKDSU\\Zhang et al. - 2022 - Unsupervised SAR and Optical Image Matching Using .pdf:application/pdf},
}
@article{xiong_confounder-free_2022,
title = {A {Confounder}-{Free} {Fusion} {Network} for {Aerial} {Image} {Scene} {Feature} {Representation}},
volume = {15},
issn = {2151-1535},
url = {https://ieeexplore.ieee.org/document/9817622},
doi = {10.1109/JSTARS.2022.3189052},
abstract = {The increasing number and complex content of aerial images have made some recent methods based on deep learning not fit well with different aerial image processing tasks. The coarse-grained feature representation proposed by these methods is not discriminative enough. Besides, the confounding factors in the datasets and long-tailed distribution of the training data will lead to biased and spurious associations among the objects of aerial images. This study proposes a confounder-free fusion network (CFF-NET) to address the challenges. Global and local feature extraction branches are designed to capture comprehensive and fine-grained deep features from the whole image. Specifically, to extract the discriminative local feature and explore the contextual information across different regions, the models based on gated recurrent units are constructed to extract features of the image region and output the important weight of each region. Furthermore, the confounder-free object feature extraction branch is proposed to generate reasonable visual attention and provide more multigrained image information. It also eliminates the spurious and biased visual relationships of the image on the object level. Finally, the output of the three branches is combined to obtain the fusion feature representation. Extensive experiments are conducted on the three popular aerial image processing tasks: 1) image classification, 2) image retrieval, and 3) image captioning. It is found that the proposed CFF-NET achieves reasonable and state-of-the-art results, including high-level tasks such as aerial image captioning.},
urldate = {2024-04-16},
journal = {IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing},
author = {Xiong, Wei and Xiong, Zhenyu and Cui, Yaqi},
year = {2022},
note = {Conference Name: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing},
keywords = {Feature extraction, Image classification, Visualization, Semantics, Task analysis, Aerial image processing, causal inference, feature representation, Image processing, Image retrieval, visual attention},
pages = {5440--5454},
file = {Xiong et al. - 2022 - A Confounder-Free Fusion Network for Aerial Image .pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\WG4W9SFR\\Xiong et al. - 2022 - A Confounder-Free Fusion Network for Aerial Image .pdf:application/pdf},
}
@article{seong_deep_2021,
title = {Deep {Spatiotemporal} {Attention} {Network} for {Fine} {Particle} {Matter} 2.5 {Concentration} {Prediction} {With} {Causality} {Analysis}},
volume = {9},
issn = {2169-3536},
url = {https://ieeexplore.ieee.org/document/9432810},
doi = {10.1109/ACCESS.2021.3080828},
abstract = {The increasing concentration of air pollutants, caused by industrialization and economic growth, is adversely affecting public health. Therefore, accurately measuring and predicting air pollution has been an important societal issue. With the era of big data and the development of artificial intelligence technologies, air pollution concentration is now being measured and recorded in real-time using different sensors. Studies have attempted to predict air pollution concentration using deep learning-based spatiotemporal prediction. This, in turn, is based on distance networks. In these studies, the distance network used to predict air pollution simply reflects the distance. However, since air pollutants cannot move over high mountain ranges and move according to the wind, the station network should include the effect of terrain and the wind direction. Previous studies do not consider these effects. To overcome these limitations, this study proposes a novel station network that combines distance and causality networks based on transfer entropy. To evaluate the performance of the proposed method, out-of-sample experiments with an hourly dataset are performed from January 2017 to October 2020 using information from 186 stations in the Republic of Korea. The results suggest that the proposed method showed state-of-the-art performance compared to existing distance-based algorithms.},
urldate = {2024-04-16},
journal = {IEEE Access},
author = {Seong, Nohyoon},
year = {2021},
note = {Conference Name: IEEE Access},
keywords = {Deep learning, deep learning, Predictive models, Time series analysis, Spatiotemporal phenomena, Pollution measurement, Air pollution, Air pollution prediction, Atmospheric modeling, causality analysis, spatiotemporal deep learning, transfer entropy},
pages = {73230--73239},
file = {Seong - 2021 - Deep Spatiotemporal Attention Network for Fine Par.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\ALUYP3UI\\Seong - 2021 - Deep Spatiotemporal Attention Network for Fine Par.pdf:application/pdf},
}
@inproceedings{gomez-chova_learning_2012,
title = {Learning with the kernel signal to noise ratio},
url = {https://ieeexplore.ieee.org/document/6349715},
doi = {10.1109/MLSP.2012.6349715},
abstract = {This paper presents the application of the kernel signal to noise ratio (KSNR) in the context of feature extraction to general machine learning and signal processing domains. The proposed approach maximizes the signal variance while minimizes the estimated noise variance in a reproducing kernel Hilbert space (RKHS). The KSNR can be used in any kernel method to deal with correlated (possibly non-Gaussian) noise. We illustrate the method in nonlinear regression examples, dependence estimation and causal inference, nonlinear channel equalization, and nonlinear feature extraction from high-dimensional satellite images. Results show that the proposed KSNR yields more fitted solutions and extracts more noise-free features when confronted with standard approaches.},
urldate = {2024-04-16},
booktitle = {2012 {IEEE} {International} {Workshop} on {Machine} {Learning} for {Signal} {Processing}},
author = {Gómez-Chova, Luis and Camps-Valls, Gustavo},
month = sep,
year = {2012},
note = {ISSN: 2378-928X},
keywords = {Estimation, Feature extraction, dependence estimation, classification, feature extraction, Hilbert space, Kernel, Kernel methods, regression, signal to noise ratio, Signal to noise ratio, Standards},
pages = {1--6},
file = {Gómez-Chova and Camps-Valls - 2012 - Learning with the kernel signal to noise ratio.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\RZZRUABY\\Gómez-Chova and Camps-Valls - 2012 - Learning with the kernel signal to noise ratio.pdf:application/pdf},
}
@inproceedings{ma_hyperspectral_2021,
title = {Hyperspectral image recognition based on lightweight causal convolutional network},
url = {https://ieeexplore.ieee.org/document/9389851},
doi = {10.1109/ICBAIE52039.2021.9389851},
abstract = {Due to the phenomenon of mixed pixels in hyperspectral images, the spatial relationship between pixels in feature recognition is more important than in common image classification. Traditional convolutional neural networks can extract spatial structure features, but it cannot accurately express the spatial causality between image pixels. A lightweight deep learning network architecture is proposed based on causal convolution, which can effectively mine the intrinsic causality between pixels in hyperspectral images. Through comparative experiments in three public datasets, it is shown that compared with the traditional convolutional neural network, this network not only maintains the accuracy in ground objects recognition but also greatly reduces the model parameters. The structure proposed in this paper has high practical value in the recognition task of hyperspectral image.},
urldate = {2024-04-16},
booktitle = {2021 {IEEE} 2nd {International} {Conference} on {Big} {Data}, {Artificial} {Intelligence} and {Internet} of {Things} {Engineering} ({ICBAIE})},
author = {Ma, Qiaoyu and Liu, Yang and Wang, Xintong and Yuan, Biao and Zhang, Kai},
month = mar,
year = {2021},
keywords = {Deep learning, Convolution, Convolutional neural networks, Feature extraction, causal convolution, hyperspectral image, Hyperspectral imaging, Image recognition, lightweight learning, Network architecture, object detection},
pages = {431--435},
file = {Ma et al. - 2021 - Hyperspectral image recognition based on lightweig.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\VG26UNIA\\Ma et al. - 2021 - Hyperspectral image recognition based on lightweig.pdf:application/pdf},
}
@article{gomez-chova_signal--noise_2018,
title = {Signal-to-noise ratio in reproducing kernel {Hilbert} spaces},
volume = {112},
issn = {0167-8655, 1872-7344},
url = {https://www.webofscience.com/api/gateway?GWVersion=2&SrcAuth=DOISource&SrcApp=WOS&KeyAID=10.1016%2Fj.patrec.2018.06.004&DestApp=DOI&SrcAppSID=EUW1ED0AD8VFWUp4rXX55OusTpL6V&SrcJTitle=PATTERN+RECOGNITION+LETTERS&DestDOIRegistrantName=Elsevier},
doi = {10.1016/j.patrec.2018.06.004},
abstract = {This paper introduces the kernel signal-to-noise ratio (kSNR) for different machine learning and signal processing applications. The kSNR seeks to maximize the signal variance while minimizing the estimated noise variance explicitly in a reproducing kernel Hilbert space (rkHs). The kSNR gives rise to considering complex signal-to-noise relations beyond additive noise models, and can be seen as a useful regularizer for feature extraction and dimensionality reduction. We show that the kSNR generalizes kernel PCA (and other spectral dimensionality reduction methods), least squares SVM, and kernel ridge regression to deal with cases where signal and noise cannot be assumed independent. We give computationally efficient alternatives based on reduced-rank Nystrom and projection on random Fourier features approximations, and analyze the bounds of performance and its stability. We illustrate the method through different examples, including nonlinear regression, nonlinear classification in channel equalization, nonlinear feature extraction from high-dimensional spectral satellite images, and bivariate causal inference. Experimental results show that the proposed kSNR yields more accurate solutions and extracts more noise-free features when compared to standard approaches. (c) 2018 Elsevier B.V. All rights reserved.},
language = {English},
urldate = {2024-04-16},
journal = {PATTERN RECOGNITION LETTERS},
author = {Gomez-Chova, Luis and Santos-Rodriguez, Raul and Camps-Valls, Gustau},
month = sep,
year = {2018},
note = {Num Pages: 8
Place: Amsterdam
Publisher: Elsevier Science Bv
Web of Science ID: WOS:000443950800011},
keywords = {Causal inference, Feature extraction, Kernel methods, Heteroscedastic, Noise model, Signal classification, Signal-to-noise ratio, SNR},
pages = {75--82},
file = {Submitted Version:C\:\\Users\\kazsa11\\Zotero\\storage\\AD5PVA9H\\Gomez-Chova et al. - 2018 - Signal-to-noise ratio in reproducing kernel Hilber.pdf:application/pdf},
}
@article{jay_alcohol_2020,
title = {Alcohol outlets and firearm violence: a place-based case-control study using satellite imagery and machine learning},
volume = {26},
issn = {1353-8047, 1475-5785},
shorttitle = {Alcohol outlets and firearm violence},
url = {https://www.webofscience.com/api/gateway?GWVersion=2&SrcAuth=DynamicDOIArticle&SrcApp=WOS&KeyAID=10.1136%2Finjuryprev-2019-043248&DestApp=DOI&SrcAppSID=EUW1ED0AD8VFWUp4rXX55OusTpL6V&SrcJTitle=INJURY+PREVENTION&DestDOIRegistrantName=BMJ},
doi = {10.1136/injuryprev-2019-043248},
abstract = {Introduction This article proposes a novel method for matching places based on visual similarity, using high-resolution satellite imagery and machine learning. This approach strengthens comparisons when the built environment is a potential confounder, as in many injury research studies. Methods As an example, I apply this method to study the spatial influence of alcohol outlets (AOs) on firearm violence in Philadelphia, Pennsylvania, specifically beer stores and bar/restaurants. Using a case-control framework, city blocks with shootings in 2017-2018 were matched with similar-looking blocks with no shootings, based on analysis with a pretrained convolutional neural network and t-distributed stochastic neighbour embedding. Logistic regression was used to estimate the OR of a shooting on the same block as an AO and within one-block and two-block distances, conditional on additional factors such as land use, demographic composition and illegal drug activity. Results The case-control matches were similar in visual appearance, on human inspection, and were well balanced on covariate measures. The fully adjusted model estimated an increased shootings risk for locations with beer stores within one block, OR=1.5, 95\% CI 1.1 to 2.1, p=0.02, and locations with bar/restaurants on the same block, OR=1.6, 95\% CI 1.1 to 2.4, p=0.02. Conclusion These findings align with previous study findings while addressing the concern that AOs might systematically be located in certain kinds of environments, providing stronger evidence of a causal effect on nearby firearm violence. Matching on visual similarity can improve observational injury studies involving place-based risks.},
language = {English},
number = {1},
urldate = {2024-04-16},
journal = {INJURY PREVENTION},
author = {Jay, Jonathan},
month = feb,
year = {2020},
note = {Num Pages: 6
Place: London
Publisher: BMJ Publishing Group
Web of Science ID: WOS:000514665600011},
keywords = {RISK, IMPACT, ASSAULT, CONSUMPTION, CRIME, STREET SEGMENTS, VACANT LAND},
pages = {61--66},
file = {Full Text:C\:\\Users\\kazsa11\\Zotero\\storage\\C3DD9NWT\\Jay - 2020 - Alcohol outlets and firearm violence a place-base.pdf:application/pdf},
}
@article{moe_increased_2021,
title = {Increased {Use} of {Bayesian} {Network} {Models} {Has} {Improved} {Environmental} {Risk} {Assessments}},
volume = {17},
issn = {1551-3777, 1551-3793},
url = {https://www.webofscience.com/api/gateway?GWVersion=2&SrcAuth=DOISource&SrcApp=WOS&KeyAID=10.1002%2Fieam.4369&DestApp=DOI&SrcAppSID=EUW1ED0AD8VFWUp4rXX55OusTpL6V&SrcJTitle=INTEGRATED+ENVIRONMENTAL+ASSESSMENT+AND+MANAGEMENT&DestDOIRegistrantName=Wiley+%28John+Wiley+%26+Sons%29},
doi = {10.1002/ieam.4369},
abstract = {Environmental and ecological risk assessments are defined as the process for evaluating the likelihood that the environment may be impacted as a result of exposure to stressors. Although this definition implies the calculation of probabilities, risk assessments traditionally rely on nonprobabilistic methods such as calculation of a risk quotient. Bayesian network (BN) models are a tool for probabilistic and causal modeling, increasingly used in many fields of environmental science. Bayesian networks are defined as directed acyclic graphs where the causal relationships and the associated uncertainty are quantified in conditional probability tables. Bayesian networks inherently incorporate uncertainty and can integrate a variety of information types, including expert elicitation. During the last 2 decades, there has been a steady increase in reports on BN applications in environmental risk assessment and management. At recent annual meetings of the Society of Environmental Toxicology and Chemistry (SETAC) North America and SETAC Europe, a number of applications of BN models were presented along with new theoretical developments. Likewise, recent meetings of the European Geosciences Union (EGU) have dedicated sessions to Bayesian modeling in relation to water quality. This special series contains 10 articles based on presentations in these sessions, reflecting a range of BN applications to systems, ranging from cells and populations to watersheds and national scale. The articles report on recent progress in many topics, including climate and management scenarios, ecological and socioeconomic endpoints, machine learning, diagnostic inference, and model evaluation. They demonstrate that BNs can be adapted to established conceptual frameworks used to support environmental risk assessments, such as adverse outcome pathways and the relative risk model. The contributions from EGU demonstrate recent advancements in areas such as spatial (geographic information system [GIS]-based) and temporal (dynamic) BN modeling. In conclusion, this special series supports the prediction that increased use of Bayesian network models will improve environmental risk assessments. Integr Environ Assess Manag 2020;00:1-9. (c) 2020 The Authors. Integrated Environmental Assessment and Management published by Wiley Periodicals LLC on behalf of Society of Environmental Toxicology \& Chemistry (SETAC)},
language = {English},
number = {1},
urldate = {2024-04-16},
journal = {INTEGRATED ENVIRONMENTAL ASSESSMENT AND MANAGEMENT},
author = {Moe, S. Jannicke and Carriger, John F. and Glendell, Miriam},
month = jan,
year = {2021},
note = {Num Pages: 9
Place: Hoboken
Publisher: Wiley
Web of Science ID: WOS:000597314900001},
keywords = {Bayesian networks, BELIEF NETWORKS, Causal modeling, Ecological risk assessment, Environmental risk assessment, FISH, MANAGEMENT, Probabilistic modeling, TOOLS},
pages = {53--61},
file = {Accepted Version:C\:\\Users\\kazsa11\\Zotero\\storage\\KTLWQZPB\\Moe et al. - 2021 - Increased Use of Bayesian Network Models Has Impro.pdf:application/pdf},
}
@article{yu_study_2021,
title = {Study becomes insight: {Ecological} learning from machine learning},
volume = {12},
issn = {2041-210X, 2041-2096},
shorttitle = {Study becomes insight},
url = {https://www.webofscience.com/api/gateway?GWVersion=2&SrcAuth=DOISource&SrcApp=WOS&KeyAID=10.1111%2F2041-210x.13686&DestApp=DOI&SrcAppSID=EUW1ED0AD8VFWUp4rXX55OusTpL6V&SrcJTitle=METHODS+IN+ECOLOGY+AND+EVOLUTION&DestDOIRegistrantName=Wiley+%28Blackwell+Publishing%29},
doi = {10.1111/2041-210X.13686},
abstract = {The ecological and environmental science communities have embraced machine learning (ML) for empirical modelling and prediction. However, going beyond prediction to draw insights into underlying functional relationships between response variables and environmental 'drivers' is less straightforward. Deriving ecological insights from fitted ML models requires techniques to extract the 'learning' hidden in the ML models. We revisit the theoretical background and effectiveness of four approaches for deriving insights from ML: ranking independent variable importance (Gini importance, GI; permutation importance, PI; split importance, SI; and conditional permutation importance, CPI), and two approaches for inference of bivariate functional relationships (partial dependence plots, PDP; and accumulated local effect plots, ALE). We also explore the use of a surrogate model for visualization and interpretation of complex multi-variate relationships between response variables and environmental drivers. We examine the challenges and opportunities for extracting ecological insights with these interpretation approaches. Specifically, we aim to improve interpretation of ML models by investigating how effectiveness relates to (a) interpretation algorithm, (b) sample size and (c) the presence of spurious explanatory variables. We base the analysis on simulations with known underlying functional relationships between response and predictor variables, with added white noise and the presence of correlated but non-influential variables. The results indicate that deriving ecological insight is strongly affected by interpretation algorithm and spurious variables, and moderately impacted by sample size. Removing spurious variables improves interpretation of ML models. Meanwhile, increasing sample size has limited value in the presence of spurious variables, but increasing sample size does improves performance once spurious variables are omitted. Among the four ranking methods, SI is slightly more effective than the other methods in the presence of spurious variables, while GI and SI yield higher accuracy when spurious variables are removed. PDP is more effective in retrieving underlying functional relationships than ALE, but its reliability declines sharply in the presence of spurious variables. Visualization and interpretation of the interactive effects of predictors and the response variable can be enhanced using surrogate models, including three-dimensional visualizations and use of loess planes to represent independent variable effects and interactions. Machine learning analysts should be aware that including correlated independent variables in ML models with no clear causal relationship to response variables can interfere with ecological inference. When ecological inference is important, ML models should be constructed with independent variables that have clear causal effects on response variables. While interpreting ML models for ecological inference remains challenging, we show that careful choice of interpretation methods, exclusion of spurious variables and adequate sample size can provide more and better opportunities to 'learn from machine learning'.},
language = {English},
number = {11},
urldate = {2024-04-16},
journal = {METHODS IN ECOLOGY AND EVOLUTION},
author = {Yu, Qiuyan and Ji, Wenjie and Prihodko, Lara and Ross, C. Wade and Anchang, Julius Y. and Hanan, Niall P.},
month = nov,
year = {2021},
note = {Num Pages: 12
Place: Hoboken
Publisher: Wiley
Web of Science ID: WOS:000681600400001},
keywords = {bivariate functional relationship, boosted regression tree (BRT), CLASSIFICATION, COVER, ecological inference, FIRE, GRADIENT, interpretation of machine learning models, random forest (RF), RANDOM FORESTS, SAMPLE-SIZE, SAVANNAS, SPECIES-DIVERSITY, TREES, variable importance},
pages = {2117--2128},
file = {Full Text:C\:\\Users\\kazsa11\\Zotero\\storage\\BGLV458V\\Yu et al. - 2021 - Study becomes insight Ecological learning from ma.pdf:application/pdf},
}
@article{tesch_variant_2021,
title = {Variant {Approach} for {Identifying} {Spurious} {Relations} {That} {Deep} {Learning} {Models} {Learn}},
volume = {3},
issn = {2624-9375},
url = {https://www.webofscience.com/api/gateway?GWVersion=2&SrcAuth=DynamicDOIArticle&SrcApp=WOS&KeyAID=10.3389%2Ffrwa.2021.745563&DestApp=DOI&SrcAppSID=EUW1ED0AD8VFWUp4rXX55OusTpL6V&SrcJTitle=FRONTIERS+IN+WATER&DestDOIRegistrantName=Frontiers+Media+SA},
doi = {10.3389/frwa.2021.745563},
abstract = {A deep learning (DL) model learns a function relating a set of input variables with a set of target variables. While the representation of this function in form of the DL model often lacks interpretability, several interpretation methods exist that provide descriptions of the function (e.g., measures of feature importance). On the one hand, these descriptions may build trust in the model or reveal its limitations. On the other hand, they may lead to new scientific understanding. In any case, a description is only useful if one is able to identify if parts of it reflect spurious instead of causal relations (e.g., random associations in the training data instead of associations due to a physical process). However, this can be challenging even for experts because, in scientific tasks, causal relations between input and target variables are often unknown or extremely complex. Commonly, this challenge is addressed by training separate instances of the considered model on random samples of the training set and identifying differences between the obtained descriptions. Here, we demonstrate that this may not be sufficient and propose to additionally consider more general modifications of the prediction task. We refer to the proposed approach as variant approach and demonstrate its usefulness and its superiority over pure sampling approaches with two illustrative prediction tasks from hydrometeorology. While being conceptually simple, to our knowledge the approach has not been formalized and systematically evaluated before.{\textless}/p{\textgreater}},
language = {English},
urldate = {2024-04-16},
journal = {FRONTIERS IN WATER},
author = {Tesch, Tobias and Kollet, Stefan and Garcke, Jochen},
month = sep,
year = {2021},
note = {Num Pages: 15
Place: Lausanne
Publisher: Frontiers Media Sa
Web of Science ID: WOS:000705052600001},
keywords = {machine learning, causality, CAUSAL INFERENCE, geoscience, hydrometeorology, INTERPRETABILITY, interpretable deep learning, NEURAL-NETWORKS, spurious correlation, statistical model},
pages = {745563},
file = {Full Text:C\:\\Users\\kazsa11\\Zotero\\storage\\Y5G9ZMBZ\\Tesch et al. - 2021 - Variant Approach for Identifying Spurious Relation.pdf:application/pdf},
}
@article{ratledge_using_2022,
title = {Using machine learning to assess the livelihood impact of electricity access},
volume = {611},
issn = {0028-0836, 1476-4687},
url = {https://www.webofscience.com/api/gateway?GWVersion=2&SrcAuth=DOISource&SrcApp=WOS&KeyAID=10.1038%2Fs41586-022-05322-8&DestApp=DOI&SrcAppSID=EUW1ED0AD8VFWUp4rXX55OusTpL6V&SrcJTitle=NATURE&DestDOIRegistrantName=Springer+Science+and+Business+Media+LLC},
doi = {10.1038/s41586-022-05322-8},
abstract = {In many regions of the world, sparse data on key economic outcomes inhibit the development, targeting and evaluation of public policy(1,2). We demonstrate how advancements in satellite imagery and machine learning (ML) can help ameliorate these data and inference challenges. In the context of an expansion of the electrical grid across Uganda, we show how a combination of satellite imagery and computer vision can be used to develop local-level livelihood measurements appropriate for inferring the causal impact of electricity access on livelihoods. We then show how ML-based inference techniques deliver more reliable estimates of the causal impact of electrification than traditional alternatives when applied to these data. We estimate that grid access improves-village-level asset wealth in rural Uganda by up to 0.15 standard deviations, more than doubling the growth rate during our study period relative to untreated areas. Our results provide country-scale evidence on the impact of grid-based infrastructure investment and our methods provide a low-cost, generalizable approach to future policy evaluation in data-sparse environments.},
language = {English},
number = {7936},
urldate = {2024-04-16},
journal = {NATURE},
author = {Ratledge, Nathan and Cadamuro, Gabe and de la Cuesta, Brandon and Stigler, Matthieu and Burke, Marshall},
month = nov,
year = {2022},
note = {Num Pages: 23
Place: Berlin
Publisher: Nature Portfolio
Web of Science ID: WOS:000884842400024},
keywords = {RURAL ELECTRIFICATION},
pages = {491--+},
file = {s41586-022-05322-8.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\I38V5SDI\\s41586-022-05322-8.pdf:application/pdf},
}
@article{chen_urban_2023,
title = {Urban {Area} {Characterization} and {Structure} {Analysis}: {A} {Combined} {Data}-{Driven} {Approach} by {Remote} {Sensing} {Information} and {Spatial}-{Temporal} {Wireless} {Data}},
volume = {15},
issn = {2072-4292},
shorttitle = {Urban {Area} {Characterization} and {Structure} {Analysis}},
url = {https://www.webofscience.com/api/gateway?GWVersion=2&SrcAuth=DOISource&SrcApp=WOS&KeyAID=10.3390%2Frs15041041&DestApp=DOI&SrcAppSID=EUW1ED0AD8VFWUp4rXX55OusTpL6V&SrcJTitle=REMOTE+SENSING&DestDOIRegistrantName=MDPI+AG},
doi = {10.3390/rs15041041},
abstract = {Analysis of urban area function is crucial for urban development. Urban area function features can help to conduct better urban planning and transportation planning. With development of urbanization, urban area function becomes complex. In order to accurately extract function features, researchers have proposed multisource data mining methods that combine urban remote sensing and other data. Therefore, the research of efficient multisource data analysis tools has become a new hot topic. In this paper, a novel urban data analysis method combining spatiotemporal wireless network data and remote sensing data was proposed. First, a Voronoi-diagram-based method was used to divide the urban remote sensing images into zones. Second, we combined period and trend components of wireless network traffic data to mine urban function structure. Third, for multisource supported urban simulation, we designed a novel spatiotemporal city computing method combining graph attention network (GAT) and gated recurrent unit (GRU) to analyze spatiotemporal urban data. The final results prove that our method performs better than other commonly used methods. In addition, we calculated the commuting index of each zone by wireless network data. Combined with the urban simulation conducted in this paper, the dynamic changes of urban area features can be sensed in advance for a better sustainable urban development.},
language = {English},
number = {4},
urldate = {2024-04-16},
journal = {REMOTE SENSING},
author = {Chen, Xiangyu and Zhang, Kaisa and Chuai, Gang and Gao, Weidong and Si, Zhiwei and Hou, Yijian and Liu, Xuewen},
month = feb,
year = {2023},
note = {Num Pages: 19
Place: Basel
Publisher: MDPI
Web of Science ID: WOS:000939969000001},
keywords = {deep learning, ENTROPY, city computing, EMPIRICAL MODE DECOMPOSITION, GRANGER CAUSALITY, NETWORK, PATTERNS, REGION, spatiotemporal big data, urban remote sensing, urban simulation, wireless traffic},
pages = {1041},
file = {Full Text:C\:\\Users\\kazsa11\\Zotero\\storage\\MUP8UJ3V\\Chen et al. - 2023 - Urban Area Characterization and Structure Analysis.pdf:application/pdf},
}
@article{runge_causal_2023,
title = {Causal inference for time series},
volume = {4},
issn = {2662-138X},
url = {https://www.webofscience.com/api/gateway?GWVersion=2&SrcAuth=DynamicDOIArticle&SrcApp=WOS&KeyAID=10.1038%2Fs43017-023-00431-y&DestApp=DOI&SrcAppSID=EUW1ED0AD8VFWUp4rXX55OusTpL6V&SrcJTitle=NATURE+REVIEWS+EARTH+%26+ENVIRONMENT&DestDOIRegistrantName=Springer+Science+and+Business+Media+LLC},
doi = {10.1038/s43017-023-00431-y},
abstract = {Many research questions in Earth and environmental sciences are inherently causal, requiring robust analyses to establish whether and how changes in one variable cause changes in another. Causal inference provides the theoretical foundations to use data and qualitative domain knowledge to quantitatively answer these questions, complementing statistics and machine learning techniques. However, there is still a broad language gap between the methodological and domain science communities. In this Technical Review, we explain the use of causal inference frameworks with a focus on the challenges of time series data. Domain-adapted explanations, method guidance and practical case studies provide an accessible summary of methods for causal discovery and causal effect estimation. Examples from climate and biogeosciences illustrate typical challenges, such as contemporaneous causation, hidden confounding and non-stationarity, and some strategies to address these challenges. Integrating causal thinking into data-driven science will facilitate process understanding and more robust machine learning and statistical models for Earth and environmental sciences, enabling the tackling of many open problems with relevant environmental, economic and societal implications. Earth sciences often investigate the causal relationships between processes and events, but there is confusion about the correct use of methods to learn these relationships from data. This Technical Review explains the application of causal inference techniques to time series and demonstrates its use through two examples of climate and biosphere-related investigations.},
language = {English},
number = {7},
urldate = {2024-04-16},
journal = {NATURE REVIEWS EARTH \& ENVIRONMENT},
author = {Runge, Jakob and Gerhardus, Andreas and Varando, Gherardo and Eyring, Veronika and Camps-Valls, Gustau},
month = jul,
year = {2023},
note = {Num Pages: 19
Place: London
Publisher: Springernature
Web of Science ID: WOS:001017712700002},
keywords = {NETWORKS, PREDICTION, DISCOVERY, IDENTIFICATION, INDEPENDENCE, INVESTIGATE, MODEL EVALUATION, NONLINEAR GRANGER-CAUSALITY, RAINFALL, TELECONNECTIONS},
pages = {487--505},
file = {Runge et al. - 2023 - Causal inference for time series.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\NL4QVMKU\\Runge et al. - 2023 - Causal inference for time series.pdf:application/pdf},
}
@article{zheng_changemask_2022,
title = {{ChangeMask}: {Deep} multi-task encoder-transformer-decoder architecture for semantic change detection},
volume = {183},
issn = {09242716},
shorttitle = {{ChangeMask}},
url = {https://linkinghub.elsevier.com/retrieve/pii/S0924271621002835},
doi = {10.1016/j.isprsjprs.2021.10.015},
abstract = {Multi-temporal high spatial resolution earth observation makes it possible to detect complex urban land surface changes, which is a significant and challenging task in remote sensing communities. Previous works mainly focus on binary change detection (BCD) based on modern technologies, e.g., deep fully convolutional network (FCN), whereas the deep network architecture for semantic change detection (SCD) is insufficiently explored in current literature. In this paper, we propose a deep multi-task encoder-transformer-decoder architecture (ChangeMask) designed by exploring two important inductive biases: sematic-change causal relationship and temporal sym metry. ChangeMask decouples the SCD into a temporal-wise semantic segmentation and a BCD, and then in tegrates these two tasks into a general encoder-transformer-decoder framework. In the encoder part, we design a semantic-aware encoder to model the semantic-change causal relationship. This encoder is only used to learn semantic representation and then learn change representation from semantic representation via a later trans former module. In this way, change representation can constrain semantic representation during training, which introduces a regularization to reduce the risk of overfitting. To learn a robust change representation from se mantic representation, we propose a temporal-symmetric transformer (TST) to guarantee temporal symmetry for change representation and keep it discriminative. Based on the above semantic representation and change representation, we adopt simple multi-task decoders to output semantic change map. Benefiting from the differentiable building blocks, ChangeMask can be trained by a multi-task loss function, which significantly simplifies the whole pipeline of applying ChangeMask. The comprehensive experimental results on two largescale SCD datasets confirm the effectiveness and superiority of ChangeMask in SCD. Besides, to demonstrate the potential value in real-world applications, e.g., automatic urban analysis and decision-making, we deploy the ChangeMask to map a large geographic area covering 30 km2 with 300 million pixels. Code will be made available.},
language = {en},
urldate = {2024-04-17},
journal = {ISPRS Journal of Photogrammetry and Remote Sensing},
author = {Zheng, Zhuo and Zhong, Yanfei and Tian, Shiqi and Ma, Ailong and Zhang, Liangpei},
month = jan,
year = {2022},
keywords = {Deep learning, Remote sensing, CLASSIFICATION, Change detection, LAND-COVER, Multi-task learning, Multi-temporal, Semantic segmentation, Temporal symmetry},
pages = {228--239},
file = {Zheng et al. - 2022 - ChangeMask Deep multi-task encoder-transformer-de.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\WNZZI47U\\Zheng et al. - 2022 - ChangeMask Deep multi-task encoder-transformer-de.pdf:application/pdf},
}
@article{mu_pirt_2023,
title = {{PIRT}: {A} {Physics}-{Informed} {Red} {Tide} {Deep} {Learning} {Forecast} {Model} {Considering} {Causal}-{Inferred} {Predictors} {Selection}},
volume = {20},
issn = {1558-0571},
shorttitle = {{PIRT}},
url = {https://ieeexplore.ieee.org/document/10056967},
doi = {10.1109/LGRS.2023.3250642},
abstract = {In this letter, a Physics-Informed Red Tide (PIRT) forecast model considering causal-inferred predictors selection is proposed. Specifically, the directed acyclic graph-graph neural network (DAG-GNN) method is first applied to quantify the causality among multiple ocean-atmosphere-biology variables for selecting the most significant predictors of the red tides (or other chlorophyll variations). Then, the encoder-decoder model consisting of an Energy Attention Module (EAM) is built for daily red tide forecasting. The multisourced multivariate dataset during 2010–2020 covering the East China Sea serves to train and evaluate PIRT. The experimental results demonstrate that the predictors in the learned causal graph are closely related to the occurrence and decay of red tides, which exhibits high physical interpretability. PIRT has a superior forecasting skill, the predictions of which are with highly consistent spatial patterns, especially in extreme events. The seven-lead-day forecast errors for chlorophyll are within 0.9 {\textbackslash}textmg{\textbackslash}cdot{\textbackslash}textm{\textasciicircum}-3 , which is much better than the other models. This also indicates that PIRT can be used as a reliable tool to study the ecology of the East China Sea.},
urldate = {2024-04-21},
journal = {IEEE Geoscience and Remote Sensing Letters},
author = {Mu, Bin and Qin, Bo and Yuan, Shijin and Wang, Xin and Chen, Yuxuan},
year = {2023},
note = {Conference Name: IEEE Geoscience and Remote Sensing Letters},
keywords = {Predictive models, Causal inference, Biological system modeling, Mathematical models, Decoding, Energy exchange, Forecasting, multivariate forecast, physics-informed neural network (PINN), red tide, Tides},
pages = {1--5},
file = {IEEE Xplore Full Text PDF:C\:\\Users\\kazsa11\\Zotero\\storage\\XUPITZJN\\Mu et al. - 2023 - PIRT A Physics-Informed Red Tide Deep Learning Fo.pdf:application/pdf},
}
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% preprints
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
@misc{wang_deeppvmap_2024,
address = {Rochester, NY},
type = {{SSRN} {Scholarly} {Paper}},
title = {Deeppvmap: {Deep} {Photovoltaic} {Map} for {Efficient} {Segmentation} of {Solar} {Panels} from {Low}-{Resolution} {Aerial} {Imagery}},
shorttitle = {Deeppvmap},
url = {https://papers.ssrn.com/abstract=4779346},
doi = {10.2139/ssrn.4779346},
abstract = {We present DeepPVMap, a Deep Learning (DL) system that detects and segments Photovoltaic (PV) panels out of low-resolution (50 cm) aerial imagery for efficient extraction and counting of PV or solar panels. PV panel detection and counting are crucial for monitoring solar energy development and policy evaluation. Recent studies have effectively utilized DL to segment PV panels in high-resolution remote sensing images. However, the acquisition of such images can be prohibitively expensive due to the extensive data needed to train DL models. To address these issues, we develop an efficient method for segmenting PV panels from low-res aerial images. We utilize the publicly available low-res aerial images of Taiwan from the National Land Surveying and Cartography Center (NLSC), greatly reducing data acquisition costs. U-Net with a pre-trained Transformer-based model is adopted as the encoder for PV panel segmentation, where the transfer learning design can effectively reduce the training requirement. Evaluations show that DeepPVMap achieves a high IoU score of 91.58\%, effectively generating a complete PV panel segmentation for the entire Taiwan island (36 K km², about 11 M tiles with 716,081 M pixels) in 10 hours. We show that our approach can accelerate the manual PV panel identification and annotation process by about 7× compared to manual annotation. This research represents a new large-scale study of automatic PV panel segmentation from low-res aerial images using advanced U-Net and Transformer models that will advance sustainable energy development.},
language = {en},
urldate = {2024-04-16},
author = {Wang, Yu-Hsiang and Cartus Bo-Xiang, You and Liao, Hsiung-Ming and Chang, Ming-Ching and Tsai, Richard},
month = mar,
year = {2024},
keywords = {Deep Learning, Semantic segmentation, Low-resolution aerial images, Renewable Energy, Solar photovoltaic panel detection},
file = {Full Text PDF:C\:\\Users\\kazsa11\\Zotero\\storage\\DB3Q3AIH\\Wang et al. - 2024 - Deeppvmap Deep Photovoltaic Map for Efficient Seg.pdf:application/pdf},
}
@misc{cohrs_causal_2024,
title = {Causal hybrid modeling with double machine learning},
url = {http://arxiv.org/abs/2402.13332},
abstract = {Hybrid modeling integrates machine learning with scientific knowledge to enhance interpretability, generalization, and adherence to natural laws. Nevertheless, equifinality and regularization biases pose challenges in hybrid modeling to achieve these purposes. This paper introduces a novel approach to estimating hybrid models via a causal inference framework, specifically employing Double Machine Learning (DML) to estimate causal effects. We showcase its use for the Earth sciences on two problems related to carbon dioxide fluxes. In the \$Q\_\{10\}\$ model, we demonstrate that DML-based hybrid modeling is superior in estimating causal parameters over end-to-end deep neural network (DNN) approaches, proving efficiency, robustness to bias from regularization methods, and circumventing equifinality. Our approach, applied to carbon flux partitioning, exhibits flexibility in accommodating heterogeneous causal effects. The study emphasizes the necessity of explicitly defining causal graphs and relationships, advocating for this as a general best practice. We encourage the continued exploration of causality in hybrid models for more interpretable and trustworthy results in knowledge-guided machine learning.},
urldate = {2024-04-16},
publisher = {arXiv},
author = {Cohrs, Kai-Hendrik and Varando, Gherardo and Carvalhais, Nuno and Reichstein, Markus and Camps-Valls, Gustau},
month = feb,
year = {2024},
note = {arXiv:2402.13332 [cs, stat]
version: 1},
keywords = {Computer Science - Machine Learning, Statistics - Methodology},
file = {arXiv.org Snapshot:C\:\\Users\\kazsa11\\Zotero\\storage\\SVHYBIKU\\2402.html:text/html;Full Text PDF:C\:\\Users\\kazsa11\\Zotero\\storage\\HHSUPV74\\Cohrs et al. - 2024 - Causal hybrid modeling with double machine learnin.pdf:application/pdf},
}
@misc{liang_quantitative_2024,
title = {Quantitative causality, causality-guided scientific discovery, and causal machine learning},
copyright = {https://creativecommons.org/licenses/by/4.0/},
url = {https://essopenarchive.org/users/551268/articles/718638-quantitative-causality-causality-guided-scientific-discovery-and-causal-machine-learning?commit=05899dab7ab628445d562aa9a300808d13acbb4d},
doi = {10.22541/essoar.170913638.81842156/v1},
abstract = {It has been said, arguably, that causality analysis should pave a promising way to interpretable deep learning and generalization. Incorporation of causality into artificial intelligence (AI) algorithms, however, is challenged with its vagueness, non-quantitiveness, computational inefficiency, etc. During the past 18 years, these challenges have been essentially resolved, with the establishment of a rigorous formalism of causality analysis initially motivated from atmospheric predictability. This not only opens a new field in the atmosphere-ocean science, namely, information flow, but also has led to scientific discoveries in other disciplines, such as quantum mechanics, neuroscience, financial economics, etc., through various applications. This note provides a brief review of the decade-long effort, including a list of major theoretical results, a sketch of the causal deep learning framework, and some representative real-world applications in geoscience pertaining to this journal, such as those on the anthropogenic cause of global warming, the decadal prediction of El Niño Modoki, the forecasting of an extreme drought in China, among others.},
language = {en},
urldate = {2024-04-16},
author = {Liang, X. San and Liang, X San and Chen, Dake and Zhang, Renhe},
month = feb,
year = {2024},
file = {Liang et al. - 2024 - Quantitative causality, causality-guided scientifi.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\KEPS66EM\\Liang et al. - 2024 - Quantitative causality, causality-guided scientifi.pdf:application/pdf},
}
@misc{sharma_domain_2024,
title = {Domain {Adaptation} for {Sustainable} {Soil} {Management} using {Causal} and {Contrastive} {Constraint} {Minimization}},
url = {http://arxiv.org/abs/2401.07175},
abstract = {Monitoring organic matter is pivotal for maintaining soil health and can help inform sustainable soil management practices. While sensor-based soil information offers higher-fidelity and reliable insights into organic matter changes, sampling and measuring sensor data is cost-prohibitive. We propose a multi-modal, scalable framework that can estimate organic matter from remote sensing data, a more readily available data source while leveraging sparse soil information for improving generalization. Using the sensor data, we preserve underlying causal relations among sensor attributes and organic matter. Simultaneously we leverage inherent structure in the data and train the model to discriminate among domains using contrastive learning. This causal and contrastive constraint minimization ensures improved generalization and adaptation to other domains. We also shed light on the interpretability of the framework by identifying attributes that are important for improving generalization. Identifying these key soil attributes that affect organic matter will aid in efforts to standardize data collection efforts.},
urldate = {2024-04-16},
publisher = {arXiv},
author = {Sharma, Somya and Sharma, Swati and Padilha, Rafael and Kiciman, Emre and Chandra, Ranveer},
month = jan,
year = {2024},
note = {arXiv:2401.07175 [cs]
version: 1},
keywords = {Computer Science - Machine Learning},
file = {arXiv.org Snapshot:C\:\\Users\\kazsa11\\Zotero\\storage\\LPP7V55M\\2401.html:text/html;Full Text PDF:C\:\\Users\\kazsa11\\Zotero\\storage\\LK6C9BLB\\Sharma et al. - 2024 - Domain Adaptation for Sustainable Soil Management .pdf:application/pdf},
}
@misc{serdavaa_satellite_2023,
address = {Rochester, NY},
type = {{SSRN} {Scholarly} {Paper}},
title = {A {Satellite} {Image} {Analysis} on {Housing} {Conditions} and the {Effectiveness} of the {Affordable} {Housing} {Mortgage} {Program} in {Mongolia}: {A} {Deep} {Learning} {Approach}},
shorttitle = {A {Satellite} {Image} {Analysis} on {Housing} {Conditions} and the {Effectiveness} of the {Affordable} {Housing} {Mortgage} {Program} in {Mongolia}},
url = {https://papers.ssrn.com/abstract=4664966},
doi = {10.2139/ssrn.4664966},
abstract = {This paper examines the effectiveness of Mongolia's subsidized public mortgage program utilizing data extracted from satellite images while demonstrating how machine learning algorithms and satellite image analysis can serve as accurate and efficient tools for monitoring and evaluating socio-economic policies on a macro-level. The paper proposes a deep-learning algorithm designed to monitor housing conditions among low-income households based on the traditional dwelling, Ger, using satellite imagery. The algorithm's accuracy is validated against annual survey data, establishing its reliability for ongoing monitoring. The paper then proceeds with the empirical analysis, revealing statistically significant cointegration and one-way causal relationships between the increased availability of subsidized mortgage loans and the decreased number of households living in poor housing conditions across areas. Specifically, a one percent increase in subsidized loans to Ger-populated areas corresponds to a 0.35 percent reduction in Ger residency. These findings underscore the subsidized mortgage program's effectiveness as a policy instrument.},
language = {en},
urldate = {2024-04-16},
author = {Serdavaa, Batkhurel},
month = dec,
year = {2023},
keywords = {Deep learning algorithm, Measuring Housing Conditions, Panel data estimation, Policy effectiveness, Satellite image analysis, Subsidized mortgage loans},
file = {Full Text PDF:C\:\\Users\\kazsa11\\Zotero\\storage\\HAAP23GX\\Serdavaa - 2023 - A Satellite Image Analysis on Housing Conditions a.pdf:application/pdf},
}
@misc{boussard_towards_2023,
title = {Towards {Causal} {Representations} of {Climate} {Model} {Data}},
url = {http://arxiv.org/abs/2312.02858},
abstract = {Climate models, such as Earth system models (ESMs), are crucial for simulating future climate change based on projected Shared Socioeconomic Pathways (SSP) greenhouse gas emissions scenarios. While ESMs are sophisticated and invaluable, machine learning-based emulators trained on existing simulation data can project additional climate scenarios much faster and are computationally efficient. However, they often lack generalizability and interpretability. This work delves into the potential of causal representation learning, specifically the {\textbackslash}emph\{Causal Discovery with Single-parent Decoding\} (CDSD) method, which could render climate model emulation efficient {\textbackslash}textit\{and\} interpretable. We evaluate CDSD on multiple climate datasets, focusing on emissions, temperature, and precipitation. Our findings shed light on the challenges, limitations, and promise of using CDSD as a stepping stone towards more interpretable and robust climate model emulation.},
urldate = {2024-04-16},
publisher = {arXiv},
author = {Boussard, Julien and Nagda, Chandni and Kaltenborn, Julia and Lange, Charlotte Emilie Elektra and Brouillard, Philippe and Gurwicz, Yaniv and Nowack, Peer and Rolnick, David},
month = dec,
year = {2023},
note = {arXiv:2312.02858 [physics, stat]
version: 2},
keywords = {Computer Science - Machine Learning, Statistics - Methodology, Computer Science - Artificial Intelligence, Physics - Atmospheric and Oceanic Physics},
file = {arXiv.org Snapshot:C\:\\Users\\kazsa11\\Zotero\\storage\\ESISE9QC\\2312.html:text/html;Full Text PDF:C\:\\Users\\kazsa11\\Zotero\\storage\\PRMDEGHH\\Boussard et al. - 2023 - Towards Causal Representations of Climate Model Da.pdf:application/pdf},
}
@misc{heuer_interpretable_2023,
title = {Interpretable multiscale {Machine} {Learning}-{Based} {Parameterizations} of {Convection} for {ICON}},
url = {http://arxiv.org/abs/2311.03251},
abstract = {In order to improve climate projections, machine learning (ML)-based parameterizations have been developed for Earth System Models (ESMs) with the goal to better represent subgrid-scale processes or to accelerate computations by emulating existent parameterizations. These data-driven models have shown some success in approximating subgrid-scale processes. However, most studies have used a particular machine learning method to parameterize the subgrid tendencies or fluxes originating from the compound effect of various small-scale processes (e.g., turbulence, radiation, convection, gravity waves) in mostly idealized settings or from superparameterizations. Here, we use a filtering technique to explicitly separate convection from these processes in data produced by the Icosahedral Non-hydrostatic modelling framework (ICON) in a realistic setting. We use a method improved by incorporating density fluctuations for computing the subgrid fluxes and compare various different machine learning algorithms to predict these fluxes. We further examine the predictions of the best performing non-deep learning model (Gradient Boosted Tree regression) and the best deep-learning model (U-Net). We discover that the U-Net learns non-causal relations between convective precipitation and convective subgrid fluxes and develop an ablated model excluding precipitating tracer species. We connect the learned relations of the U-Net to physical processes in contrast to non-deep learning-based algorithms. The ML schemes are coupled online to the host ICON model and the non-causal links reveal weaknesses in stability and precipitation predictions. Predicted precipitation extremes of the ablated U-Net show higher accuracy over the conventional convection parameterization. Thus, our results provide a significant advance upon existing ML subgrid representation in ESMs.},
urldate = {2024-04-16},
publisher = {arXiv},
author = {Heuer, Helge and Schwabe, Mierk and Gentine, Pierre and Giorgetta, Marco A. and Eyring, Veronika},
month = nov,
year = {2023},
note = {arXiv:2311.03251 [physics]
version: 1},
keywords = {Physics - Atmospheric and Oceanic Physics},
file = {arXiv.org Snapshot:C\:\\Users\\kazsa11\\Zotero\\storage\\QJ5IVAGM\\2311.html:text/html;Full Text PDF:C\:\\Users\\kazsa11\\Zotero\\storage\\FKR8PSE5\\Heuer et al. - 2023 - Interpretable multiscale Machine Learning-Based Pa.pdf:application/pdf},
}
@misc{xu_physics-aware_2024,
title = {Physics-aware {Machine} {Learning} {Revolutionizes} {Scientific} {Paradigm} for {Machine} {Learning} and {Process}-based {Hydrology}},
url = {http://arxiv.org/abs/2310.05227},
abstract = {Accurate hydrological understanding and water cycle prediction are crucial for addressing scientific and societal challenges associated with the management of water resources, particularly under the dynamic influence of anthropogenic climate change. Existing reviews predominantly concentrate on the development of machine learning (ML) in this field, yet there is a clear distinction between hydrology and ML as separate paradigms. Here, we introduce physics-aware ML as a transformative approach to overcome the perceived barrier and revolutionize both fields. Specifically, we present a comprehensive review of the physics-aware ML methods, building a structured community (PaML) of existing methodologies that integrate prior physical knowledge or physics-based modeling into ML. We systematically analyze these PaML methodologies with respect to four aspects: physical data-guided ML, physics-informed ML, physics-embedded ML, and physics-aware hybrid learning. PaML facilitates ML-aided hypotheses, accelerating insights from big data and fostering scientific discoveries. We first conduct a systematic review of hydrology in PaML, including rainfall-runoff hydrological processes and hydrodynamic processes, and highlight the most promising and challenging directions for different objectives and PaML methods. Finally, a new PaML-based hydrology platform, termed HydroPML, is released as a foundation for hydrological applications. HydroPML enhances the explainability and causality of ML and lays the groundwork for the digital water cycle's realization. The HydroPML platform is publicly available at https://hydropml.github.io/.},
urldate = {2024-04-16},
publisher = {arXiv},
author = {Xu, Qingsong and Shi, Yilei and Bamber, Jonathan and Tuo, Ye and Ludwig, Ralf and Zhu, Xiao Xiang},
month = feb,
year = {2024},
note = {arXiv:2310.05227 [physics]
version: 3},
keywords = {Computer Science - Machine Learning, Computer Science - Artificial Intelligence, Physics - Fluid Dynamics},
file = {arXiv.org Snapshot:C\:\\Users\\kazsa11\\Zotero\\storage\\D6N8EG2K\\2310.html:text/html;Full Text PDF:C\:\\Users\\kazsa11\\Zotero\\storage\\H9VIN929\\Xu et al. - 2024 - Physics-aware Machine Learning Revolutionizes Scie.pdf:application/pdf},
}
@misc{wang_causality-informed_2023,
title = {Causality-informed {Rapid} {Post}-hurricane {Building} {Damage} {Detection} in {Large} {Scale} from {InSAR} {Imagery}},
url = {http://arxiv.org/abs/2310.01565},
abstract = {Timely and accurate assessment of hurricane-induced building damage is crucial for effective post-hurricane response and recovery efforts. Recently, remote sensing technologies provide large-scale optical or Interferometric Synthetic Aperture Radar (InSAR) imagery data immediately after a disastrous event, which can be readily used to conduct rapid building damage assessment. Compared to optical satellite imageries, the Synthetic Aperture Radar can penetrate cloud cover and provide more complete spatial coverage of damaged zones in various weather conditions. However, these InSAR imageries often contain highly noisy and mixed signals induced by co-occurring or co-located building damage, flood, flood/wind-induced vegetation changes, as well as anthropogenic activities, making it challenging to extract accurate building damage information. In this paper, we introduced an approach for rapid post-hurricane building damage detection from InSAR imagery. This approach encoded complex causal dependencies among wind, flood, building damage, and InSAR imagery using a holistic causal Bayesian network. Based on the causal Bayesian network, we further jointly inferred the large-scale unobserved building damage by fusing the information from InSAR imagery with prior physical models of flood and wind, without the need for ground truth labels. Furthermore, we validated our estimation results in a real-world devastating hurricane -- the 2022 Hurricane Ian. We gathered and annotated building damage ground truth data in Lee County, Florida, and compared the introduced method's estimation results with the ground truth and benchmarked it against state-of-the-art models to assess the effectiveness of our proposed method. Results show that our method achieves rapid and accurate detection of building damage, with significantly reduced processing time compared to traditional manual inspection methods.},
urldate = {2024-04-16},
publisher = {arXiv},
author = {Wang, Chenguang and Liu, Yepeng and Zhang, Xiaojian and Li, Xuechun and Paramygin, Vladimir and Subgranon, Arthriya and Sheng, Peter and Zhao, Xilei and Xu, Susu},
month = oct,
year = {2023},
note = {arXiv:2310.01565 [cs, eess]
version: 1},
keywords = {Computer Science - Machine Learning, Computer Science - Information Retrieval, Electrical Engineering and Systems Science - Image and Video Processing},
file = {arXiv.org Snapshot:C\:\\Users\\kazsa11\\Zotero\\storage\\THDD92TY\\2310.html:text/html;Full Text PDF:C\:\\Users\\kazsa11\\Zotero\\storage\\XPKQW75G\\Wang et al. - 2023 - Causality-informed Rapid Post-hurricane Building D.pdf:application/pdf},
}
@misc{jerzak_causalimages_2023,
title = {{CausalImages}: {An} {R} {Package} for {Causal} {Inference} with {Earth} {Observation}, {Bio}-medical, and {Social} {Science} {Images}},
shorttitle = {{CausalImages}},
url = {http://arxiv.org/abs/2310.00233},
abstract = {The causalimages R package enables causal inference with image and image sequence data, providing new tools for integrating novel data sources like satellite and bio-medical imagery into the study of cause and effect. One set of functions enables image-based causal inference analyses. For example, one key function decomposes treatment effect heterogeneity by images using an interpretable Bayesian framework. This allows for determining which types of images or image sequences are most responsive to interventions. A second modeling function allows researchers to control for confounding using images. The package also allows investigators to produce embeddings that serve as vector summaries of the image or video content. Finally, infrastructural functions are also provided, such as tools for writing large-scale image and image sequence data as sequentialized byte strings for more rapid image analysis. causalimages therefore opens new capabilities for causal inference in R, letting researchers use informative imagery in substantive analyses in a fast and accessible manner.},
urldate = {2024-04-16},
publisher = {arXiv},
author = {Jerzak, Connor T. and Daoud, Adel},
month = nov,
year = {2023},
note = {arXiv:2310.00233 [cs, stat]
version: 3},
keywords = {Computer Science - Machine Learning, Statistics - Methodology, 62-07, 68U10, I.4},
file = {arXiv.org Snapshot:C\:\\Users\\kazsa11\\Zotero\\storage\\3UJ84W6Z\\2310.html:text/html;Full Text PDF:C\:\\Users\\kazsa11\\Zotero\\storage\\YGKZEL4Q\\Jerzak and Daoud - 2023 - CausalImages An R Package for Causal Inference wi.pdf:application/pdf},
}
@misc{elvira_graphs_2023,
title = {Graphs in {State}-{Space} {Models} for {Granger} {Causality} in {Climate} {Science}},
url = {http://arxiv.org/abs/2307.10703},
abstract = {Granger causality (GC) is often considered not an actual form of causality. Still, it is arguably the most widely used method to assess the predictability of a time series from another one. Granger causality has been widely used in many applied disciplines, from neuroscience and econometrics to Earth sciences. We revisit GC under a graphical perspective of state-space models. For that, we use GraphEM, a recently presented expectation-maximisation algorithm for estimating the linear matrix operator in the state equation of a linear-Gaussian state-space model. Lasso regularisation is included in the M-step, which is solved using a proximal splitting Douglas-Rachford algorithm. Experiments in toy examples and challenging climate problems illustrate the benefits of the proposed model and inference technique over standard Granger causality methods.},
urldate = {2024-04-16},
publisher = {arXiv},
author = {Elvira, Víctor and Chouzenoux, Émilie and Cerdà, Jordi and Camps-Valls, Gustau},
month = jul,
year = {2023},
note = {arXiv:2307.10703 [cs]
version: 1},
keywords = {Computer Science - Machine Learning},
file = {arXiv.org Snapshot:C\:\\Users\\kazsa11\\Zotero\\storage\\38S7QSD7\\2307.html:text/html;Full Text PDF:C\:\\Users\\kazsa11\\Zotero\\storage\\RYX2U8FT\\Elvira et al. - 2023 - Graphs in State-Space Models for Granger Causality.pdf:application/pdf},
}
@misc{giannarakis_understanding_2023,
title = {Understanding the impacts of crop diversification in the context of climate change: a machine learning approach},
shorttitle = {Understanding the impacts of crop diversification in the context of climate change},
url = {http://arxiv.org/abs/2307.08617},
abstract = {The concept of sustainable intensification in agriculture necessitates the implementation of management practices that prioritize sustainability without compromising productivity. However, the effects of such practices are known to depend on environmental conditions, and are therefore expected to change as a result of a changing climate. We study the impact of crop diversification on productivity in the context of climate change. We leverage heterogeneous Earth Observation data and contribute a data-driven approach based on causal machine learning for understanding how crop diversification impacts may change in the future. We apply this method to the country of Cyprus throughout a 4-year period. We find that, on average, crop diversification significantly benefited the net primary productivity of crops, increasing it by 2.8\%. The effect generally synergized well with higher maximum temperatures and lower soil moistures. In a warmer and more drought-prone climate, we conclude that crop diversification exhibits promising adaptation potential and is thus a sensible policy choice with regards to agricultural productivity for present and future.},
urldate = {2024-04-16},
publisher = {arXiv},
author = {Giannarakis, Georgios and Tsoumas, Ilias and Neophytides, Stelios and Papoutsa, Christiana and Kontoes, Charalampos and Hadjimitsis, Diofantos},
month = jul,
year = {2023},
note = {arXiv:2307.08617 [cs, q-bio]
version: 1},
keywords = {Computer Science - Machine Learning, Quantitative Biology - Populations and Evolution},
file = {arXiv.org Snapshot:C\:\\Users\\kazsa11\\Zotero\\storage\\UAW788DS\\2307.html:text/html;Full Text PDF:C\:\\Users\\kazsa11\\Zotero\\storage\\LJNVGF4C\\Giannarakis et al. - 2023 - Understanding the impacts of crop diversification .pdf:application/pdf},
}
@misc{debeire_bootstrap_2024,
title = {Bootstrap aggregation and confidence measures to improve time series causal discovery},
url = {http://arxiv.org/abs/2306.08946},
abstract = {Learning causal graphs from multivariate time series is a ubiquitous challenge in all application domains dealing with time-dependent systems, such as in Earth sciences, biology, or engineering, to name a few. Recent developments for this causal discovery learning task have shown considerable skill, notably the specific time-series adaptations of the popular conditional independence-based learning framework. However, uncertainty estimation is challenging for conditional independence-based methods. Here, we introduce a novel bootstrap approach designed for time series causal discovery that preserves the temporal dependencies and lag structure. It can be combined with a range of time series causal discovery methods and provides a measure of confidence for the links of the time series graphs. Furthermore, next to confidence estimation, an aggregation, also called bagging, of the bootstrapped graphs by majority voting results in bagged causal discovery methods. In this work, we combine this approach with the state-of-the-art conditional-independence-based algorithm PCMCI+. With extensive numerical experiments we empirically demonstrate that, in addition to providing confidence measures for links, Bagged-PCMCI+ improves in precision and recall as compared to its base algorithm PCMCI+, at the cost of higher computational demands. These statistical performance improvements are especially pronounced in the more challenging settings (short time sample size, large number of variables, high autocorrelation). Our bootstrap approach can also be combined with other time series causal discovery algorithms and can be of considerable use in many real-world applications.},
urldate = {2024-04-16},
publisher = {arXiv},
author = {Debeire, Kevin and Runge, Jakob and Gerhardus, Andreas and Eyring, Veronika},
month = feb,
year = {2024},
note = {arXiv:2306.08946 [stat]
version: 2},
keywords = {Statistics - Methodology, Statistics - Machine Learning},
file = {arXiv.org Snapshot:C\:\\Users\\kazsa11\\Zotero\\storage\\M7LIUXRS\\2306.html:text/html;Full Text PDF:C\:\\Users\\kazsa11\\Zotero\\storage\\CIK4WLIP\\Debeire et al. - 2024 - Bootstrap aggregation and confidence measures to i.pdf:application/pdf},
}
@misc{chouzenoux_sparse_2023,
title = {Sparse {Graphical} {Linear} {Dynamical} {Systems}},
url = {http://arxiv.org/abs/2307.03210},
abstract = {Time-series datasets are central in numerous fields of science and engineering, such as biomedicine, Earth observation, and network analysis. Extensive research exists on state-space models (SSMs), which are powerful mathematical tools that allow for probabilistic and interpretable learning on time series. Estimating the model parameters in SSMs is arguably one of the most complicated tasks, and the inclusion of prior knowledge is known to both ease the interpretation but also to complicate the inferential tasks. Very recent works have attempted to incorporate a graphical perspective on some of those model parameters, but they present notable limitations that this work addresses. More generally, existing graphical modeling tools are designed to incorporate either static information, focusing on statistical dependencies among independent random variables (e.g., graphical Lasso approach), or dynamic information, emphasizing causal relationships among time series samples (e.g., graphical Granger approaches). However, there are no joint approaches combining static and dynamic graphical modeling within the context of SSMs. This work proposes a novel approach to fill this gap by introducing a joint graphical modeling framework that bridges the static graphical Lasso model and a causal-based graphical approach for the linear-Gaussian SSM. We present DGLASSO (Dynamic Graphical Lasso), a new inference method within this framework that implements an efficient block alternating majorization-minimization algorithm. The algorithm's convergence is established by departing from modern tools from nonlinear analysis. Experimental validation on synthetic and real weather variability data showcases the effectiveness of the proposed model and inference algorithm.},
urldate = {2024-04-16},
publisher = {arXiv},
author = {Chouzenoux, Emilie and Elvira, Victor},
month = jul,
year = {2023},
note = {arXiv:2307.03210 [cs, math, stat]
version: 1},
keywords = {Computer Science - Machine Learning, Mathematics - Optimization and Control, Statistics - Computation},
file = {arXiv.org Snapshot:C\:\\Users\\kazsa11\\Zotero\\storage\\APMZF696\\2307.html:text/html;Full Text PDF:C\:\\Users\\kazsa11\\Zotero\\storage\\5CUTH48W\\Chouzenoux and Elvira - 2023 - Sparse Graphical Linear Dynamical Systems.pdf:application/pdf},
}
@misc{berrevoets_causal_2024,
title = {Causal {Deep} {Learning}},
url = {http://arxiv.org/abs/2303.02186},
abstract = {Causality has the potential to truly transform the way we solve a large number of real-world problems. Yet, so far, its potential largely remains to be unlocked as causality often requires crucial assumptions which cannot be tested in practice. To address this challenge, we propose a new way of thinking about causality -- we call this causal deep learning. Our causal deep learning framework spans three dimensions: (1) a structural dimension, which incorporates partial yet testable causal knowledge rather than assuming either complete or no causal knowledge among the variables of interest; (2) a parametric dimension, which encompasses parametric forms that capture the type of relationships among the variables of interest; and (3) a temporal dimension, which captures exposure times or how the variables of interest interact (possibly causally) over time. Causal deep learning enables us to make progress on a variety of real-world problems by leveraging partial causal knowledge (including independencies among variables) and quantitatively characterising causal relationships among variables of interest (possibly over time). Our framework clearly identifies which assumptions are testable and which ones are not, such that the resulting solutions can be judiciously adopted in practice. Using our formulation we can combine or chain together causal representations to solve specific problems without losing track of which assumptions are required to build these solutions, pushing real-world impact in healthcare, economics and business, environmental sciences and education, through causal deep learning.},
urldate = {2024-04-16},
publisher = {arXiv},
author = {Berrevoets, Jeroen and Kacprzyk, Krzysztof and Qian, Zhaozhi and van der Schaar, Mihaela},
month = feb,
year = {2024},
note = {arXiv:2303.02186 [cs]
version: 2},
keywords = {Computer Science - Machine Learning, Computer Science - Artificial Intelligence},
file = {arXiv.org Snapshot:C\:\\Users\\kazsa11\\Zotero\\storage\\VAYFCSTJ\\2303.html:text/html;Full Text PDF:C\:\\Users\\kazsa11\\Zotero\\storage\\82CCALP7\\Berrevoets et al. - 2024 - Causal Deep Learning.pdf:application/pdf},
}
@misc{jerzak_integrating_2023,
title = {Integrating {Earth} {Observation} {Data} into {Causal} {Inference}: {Challenges} and {Opportunities}},
shorttitle = {Integrating {Earth} {Observation} {Data} into {Causal} {Inference}},
url = {http://arxiv.org/abs/2301.12985},
abstract = {Observational studies require adjustment for confounding factors that are correlated with both the treatment and outcome. In the setting where the observed variables are tabular quantities such as average income in a neighborhood, tools have been developed for addressing such confounding. However, in many parts of the developing world, features about local communities may be scarce. In this context, satellite imagery can play an important role, serving as a proxy for the confounding variables otherwise unobserved. In this paper, we study confounder adjustment in this non-tabular setting, where patterns or objects found in satellite images contribute to the confounder bias. Using the evaluation of anti-poverty aid programs in Africa as our running example, we formalize the challenge of performing causal adjustment with such unstructured data -- what conditions are sufficient to identify causal effects, how to perform estimation, and how to quantify the ways in which certain aspects of the unstructured image object are most predictive of the treatment decision. Via simulation, we also explore the sensitivity of satellite image-based observational inference to image resolution and to misspecification of the image-associated confounder. Finally, we apply these tools in estimating the effect of anti-poverty interventions in African communities from satellite imagery.},
urldate = {2024-04-16},
publisher = {arXiv},
author = {Jerzak, Connor T. and Johansson, Fredrik and Daoud, Adel},
month = jan,
year = {2023},
note = {arXiv:2301.12985 [cs, stat]
version: 1},
keywords = {Computer Science - Machine Learning, Statistics - Machine Learning, 62D20, J.4, Statistics - Applications},
file = {arXiv.org Snapshot:C\:\\Users\\kazsa11\\Zotero\\storage\\JNFKTF4S\\2301.html:text/html;Full Text PDF:C\:\\Users\\kazsa11\\Zotero\\storage\\ILZBJNNX\\Jerzak et al. - 2023 - Integrating Earth Observation Data into Causal Inf.pdf:application/pdf},
}
@article{eldhose_robust_nodate,
title = {Robust {Causality} and {False} {Attribution} in {Data}-{Driven} {Earth} {Science} {Discoveries}},
abstract = {Causal and attribution studies are essential for earth scientific discoveries and critical for informing climate, ecology, and water policies. However, the current generation of methods needs to keep pace with the complexity of scientific and stakeholder challenges and data availability combined with the adequacy of data-driven methods. Unless carefully informed by physics, they run the risk of conflating correlation with causation or getting overwhelmed by estimation inaccuracies. Given that natural experiments, controlled trials, interventions, and counterfactual examinations are often impractical, information-theoretic methods have been developed and are being continually refined in the earth sciences. Here we show that transfer entropy-based causal graphs, which have recently become popular in the earth sciences with high-profile discoveries, can be spurious even when augmented with statistical significance. We develop a subsample-based ensemble approach for robust causality analysis. Simulated data, and observations in climate and ecohydrology, suggest the robustness and consistency of this approach.},
language = {en},
author = {Eldhose, Elizabeth and Chauhan, Tejasvi and Chandel, Vikram and Ghosh, Subimal and Ganguly, Auroop R},
file = {Eldhose et al. - Robust Causality and False Attribution in Data-Dri.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\5ATFFY4K\\Eldhose et al. - Robust Causality and False Attribution in Data-Dri.pdf:application/pdf},
}
@misc{biswas_trustworthy_2022,
title = {Trustworthy modelling of atmospheric formaldehyde powered by deep learning},
url = {http://arxiv.org/abs/2209.07414},
abstract = {Formaldehyde (HCHO) is one one of the most important trace gas in the atmosphere, as it is a pollutant causing respiratory and other diseases. It is also a precursor of tropospheric ozone which damages crops and deteriorates human health. Study of HCHO chemistry and long-term monitoring using satellite data is important from the perspective of human health, food security and air pollution. Dynamic atmospheric chemistry models struggle to simulate atmospheric formaldehyde and often overestimate by up to two times relative to satellite observations and reanalysis. Spatial distribution of modelled HCHO also fail to match satellite observations. Here, we present deep learning approach using a simple super-resolution based convolutional neural network towards simulating fast and reliable atmospheric HCHO. Our approach is an indirect method of HCHO estimation without the need to chemical equations. We find that deep learning outperforms dynamical model simulations which involves complicated atmospheric chemistry representation. Causality establishing the nonlinear relationships of different variables to target formaldehyde is established in our approach by using a variety of precursors from meteorology and chemical reanalysis to target OMI AURA satellite based HCHO predictions. We choose South Asia for testing our implementation as it doesnt have in situ measurements of formaldehyde and there is a need for improved quality data over the region. Moreover, there are spatial and temporal data gaps in the satellite product which can be removed by trustworthy modelling of atmospheric formaldehyde. This study is a novel attempt using computer vision for trustworthy modelling of formaldehyde from remote sensing can lead to cascading societal benefits.},
urldate = {2024-04-16},
publisher = {arXiv},
author = {Biswas, Mriganka Sekhar and Singh, Manmeet},
month = aug,
year = {2022},
note = {arXiv:2209.07414 [astro-ph, physics:physics]
version: 1},
keywords = {Computer Science - Machine Learning, Physics - Atmospheric and Oceanic Physics, Astrophysics - Earth and Planetary Astrophysics, Computer Science - Computer Vision and Pattern Recognition, Physics - Chemical Physics},
file = {arXiv.org Snapshot:C\:\\Users\\kazsa11\\Zotero\\storage\\L23B7TD6\\2209.html:text/html;Full Text PDF:C\:\\Users\\kazsa11\\Zotero\\storage\\FWFHL6HK\\Biswas and Singh - 2022 - Trustworthy modelling of atmospheric formaldehyde .pdf:application/pdf},
}
@misc{fernandez-loria_causal_2024,
title = {Causal {Scoring}: {A} {Framework} for {Effect} {Estimation}, {Effect} {Ordering}, and {Effect} {Classification}},
shorttitle = {Causal {Scoring}},
url = {http://arxiv.org/abs/2206.12532},
abstract = {This paper introduces causal scoring as a novel approach to frame causal estimation in the context of decision making. Causal scoring entails the estimation of scores that support decision making by providing insights into causal effects. We present three valuable causal interpretations of these scores: effect estimation (EE), effect ordering (EO), and effect classification (EC). In the EE interpretation, the causal score represents the effect itself. The EO interpretation implies that the score can serve as a proxy for the magnitude of the effect, enabling the sorting of individuals based on their causal effects. The EC interpretation enables the classification of individuals into high- and low-effect categories using a predefined threshold. We demonstrate the value of these alternative causal interpretations (EO and EC) through two key results. First, we show that aligning the statistical modeling with the desired causal interpretation improves the accuracy of causal estimation. Second, we establish that more flexible causal interpretations are plausible in a wider range of settings and propose conditions to assess their validity. We showcase the practical utility of causal scoring through diverse scenarios, including situations involving unobserved confounding due to self-selection, lack of data on the primary outcome of interest, or lack of data on how individuals behave when intervened. These examples illustrate how causal scoring facilitates reasoning about flexible causal interpretations of statistical estimates in various contexts. They encompass confounded estimates, effect estimates on surrogate outcomes, and even predictions about non-causal quantities as potential causal scores.},
urldate = {2024-04-16},
publisher = {arXiv},
author = {Fernández-Loría, Carlos and Loría, Jorge},
month = feb,
year = {2024},
note = {arXiv:2206.12532 [cs, stat]
version: 4},
keywords = {Computer Science - Machine Learning, Statistics - Machine Learning},
file = {arXiv.org Snapshot:C\:\\Users\\kazsa11\\Zotero\\storage\\27GVE8D9\\2206.html:text/html;Full Text PDF:C\:\\Users\\kazsa11\\Zotero\\storage\\5FBL5Q5V\\Fernández-Loría and Loría - 2024 - Causal Scoring A Framework for Effect Estimation,.pdf:application/pdf},
}
@misc{jerzak_estimating_2023,
title = {Estimating {Causal} {Effects} {Under} {Image} {Confounding} {Bias} with an {Application} to {Poverty} in {Africa}},
url = {http://arxiv.org/abs/2206.06410},
abstract = {Observational studies of causal effects require adjustment for confounding factors. In the tabular setting, where these factors are well-defined, separate random variables, the effect of confounding is well understood. However, in public policy, ecology, and in medicine, decisions are often made in non-tabular settings, informed by patterns or objects detected in images (e.g., maps, satellite or tomography imagery). Using such imagery for causal inference presents an opportunity because objects in the image may be related to the treatment and outcome of interest. In these cases, we rely on the images to adjust for confounding but observed data do not directly label the existence of the important objects. Motivated by real-world applications, we formalize this challenge, how it can be handled, and what conditions are sufficient to identify and estimate causal effects. We analyze finite-sample performance using simulation experiments, estimating effects using a propensity adjustment algorithm that employs a machine learning model to estimate the image confounding. Our experiments also examine sensitivity to misspecification of the image pattern mechanism. Finally, we use our methodology to estimate the effects of policy interventions on poverty in African communities from satellite imagery.},
urldate = {2024-04-16},
publisher = {arXiv},
author = {Jerzak, Connor T. and Johansson, Fredrik and Daoud, Adel},
month = feb,
year = {2023},
note = {arXiv:2206.06410 [cs, stat]
version: 3},
keywords = {Computer Science - Machine Learning, Statistics - Methodology, 62D20, I.2.0, I.4.0},
file = {arXiv.org Snapshot:C\:\\Users\\kazsa11\\Zotero\\storage\\7DTTLB3T\\2206.html:text/html;Full Text PDF:C\:\\Users\\kazsa11\\Zotero\\storage\\EU4ALCL4\\Jerzak et al. - 2023 - Estimating Causal Effects Under Image Confounding .pdf:application/pdf},
}
@misc{go_use_2022,
address = {Rochester, NY},
type = {{SSRN} {Scholarly} {Paper}},
title = {On the {Use} of {Satellite}-{Based} {Vehicle} {Flows} {Data} to {Assess} {Local} {Economic} {Activity}: {The} {Case} of {Philippine} {Cities}},
shorttitle = {On the {Use} of {Satellite}-{Based} {Vehicle} {Flows} {Data} to {Assess} {Local} {Economic} {Activity}},
url = {https://papers.ssrn.com/abstract=4057690},
doi = {10.2139/ssrn.4057690},
abstract = {The lack of suitable data is a key challenge in ex-post policy evaluations. This paper proposes a novel data to measure local economic activities using vehicle counts in each 500 meter (m) x 500 m tile. The metric is derived from high resolution satellite images using a machine learning algorithm. Using the opening of the new international airport terminal in Cebu, Philippines, as a quasi-experiment, we estimate the impact of the new infrastructure on the local economy of Metro Cebu. Results of the difference-in-differences analysis show that the new terminal significantly increased vehicle traffic in urban Cebu. The effect decays with distance from the airport, is stronger in areas where hotels are located, and is most pronounced in the peak months for international tourists. These findings imply that the opening of the new international terminal has enhanced Cebu's local economy through international tourism.},
language = {en},
urldate = {2024-04-16},
author = {Go, Eugenia and Nakajima, Kentaro and Sawada, Yasuyuki and Taniguchi, Kiyoshi},
month = mar,
year = {2022},
keywords = {satellite imagery data, transportation infrastructure},
file = {Full Text PDF:C\:\\Users\\kazsa11\\Zotero\\storage\\58Z56QLS\\Go et al. - 2022 - On the Use of Satellite-Based Vehicle Flows Data t.pdf:application/pdf},
}
@misc{xu_ill-posed_2022,
title = {Ill-posed {Surface} {Emissivity} {Retrieval} from {Multi}-{Geometry} {Hyperspectral} {Images} using a {Hybrid} {Deep} {Neural} {Network}},
url = {http://arxiv.org/abs/2107.04631},
abstract = {Atmospheric correction is a fundamental task in remote sensing because observations are taken either of the atmosphere or looking through the atmosphere. Atmospheric correction errors can significantly alter the spectral signature of the observations, and lead to invalid classifications or target detection. This is even more crucial when working with hyperspectral data, where a precise measurement of spectral properties is required. State-of-the-art physics-based atmospheric correction approaches require extensive prior knowledge about sensor characteristics, collection geometry, and environmental characteristics of the scene being collected. These approaches are computationally expensive, prone to inaccuracy due to lack of sufficient environmental and collection information, and often impossible for real-time applications. In this paper, a geometry-dependent hybrid neural network is proposed for automatic atmospheric correction using multi-scan hyperspectral data collected from different geometries. The proposed network can characterize the atmosphere without any additional meteorological data. A grid-search method is also proposed to solve the temperature emissivity separation problem. Results show that the proposed network has the capacity to accurately characterize the atmosphere and estimate target emissivity spectra with a Mean Absolute Error (MAE) under 0.02 for 29 different materials. This solution can lead to accurate atmospheric correction to improve target detection for real time applications.},
urldate = {2024-04-16},
publisher = {arXiv},
author = {Xu, Fangcao and Sun, Jian and Cervone, Guido and Salvador, Mark},
month = mar,
year = {2022},
note = {arXiv:2107.04631 [cs, eess]
version: 3},
keywords = {Computer Science - Machine Learning, Electrical Engineering and Systems Science - Image and Video Processing},
file = {arXiv.org Snapshot:C\:\\Users\\kazsa11\\Zotero\\storage\\INXUXZYT\\2107.html:text/html;Xu et al. - 2022 - Ill-posed Surface Emissivity Retrieval from Multi-.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\CFSFC6LD\\Xu et al. - 2022 - Ill-posed Surface Emissivity Retrieval from Multi-.pdf:application/pdf},
}
@misc{mateo-sanchis_warped_2020,
title = {Warped {Gaussian} {Processes} in {Remote} {Sensing} {Parameter} {Estimation} and {Causal} {Inference}},
url = {http://arxiv.org/abs/2012.12105},
abstract = {This paper introduces warped Gaussian processes (WGP) regression in remote sensing applications. WGP models output observations as a parametric nonlinear transformation of a GP. The parameters of such prior model are then learned via standard maximum likelihood. We show the good performance of the proposed model for the estimation of oceanic chlorophyll content from multispectral data, vegetation parameters (chlorophyll, leaf area index, and fractional vegetation cover) from hyperspectral data, and in the detection of the causal direction in a collection of 28 bivariate geoscience and remote sensing causal problems. The model consistently performs better than the standard GP and the more advanced heteroscedastic GP model, both in terms of accuracy and more sensible confidence intervals.},
urldate = {2024-04-16},
publisher = {arXiv},
author = {Mateo-Sanchis, Anna and Muñoz-Marí, Jordi and Pérez-Suay, Adrián and Camps-Valls, Gustau},
month = dec,
year = {2020},
note = {arXiv:2012.12105 [cs]},
keywords = {Computer Science - Machine Learning, Computer Science - Computer Vision and Pattern Recognition},
file = {arXiv.org Snapshot:C\:\\Users\\kazsa11\\Zotero\\storage\\6MLU9DSJ\\2012.html:text/html;Full Text PDF:C\:\\Users\\kazsa11\\Zotero\\storage\\RDFIZIWQ\\Mateo-Sanchis et al. - 2020 - Warped Gaussian Processes in Remote Sensing Parame.pdf:application/pdf},
}
@misc{morata-dolz_understanding_2020,
title = {Understanding {Climate} {Impacts} on {Vegetation} with {Gaussian} {Processes} in {Granger} {Causality}},
url = {http://arxiv.org/abs/2012.03338},
abstract = {Global warming is leading to unprecedented changes in our planet, with great societal, economical and environmental implications, especially with the growing demand of biofuels and food. Assessing the impact of climate on vegetation is of pressing need. We approached the attribution problem with a novel nonlinear Granger causal (GC) methodology and used a large data archive of remote sensing satellite products, environmental and climatic variables spatio-temporally gridded over more than 30 years. We generalize kernel Granger causality by considering the variables cross-relations explicitly in Hilbert spaces, and use the covariance in Gaussian processes. The method generalizes the linear and kernel GC methods, and comes with tighter bounds of performance based on Rademacher complexity. Spatially-explicit global Granger footprints of precipitation and soil moisture on vegetation greenness are identified more sharply than previous GC methods.},
urldate = {2024-04-16},
publisher = {arXiv},
author = {Morata-Dolz, Miguel and Bueso, Diego and Piles, Maria and Camps-Valls, Gustau},
month = dec,
year = {2020},
note = {arXiv:2012.03338 [physics]},
keywords = {Computer Science - Machine Learning, Physics - Atmospheric and Oceanic Physics},
file = {arXiv.org Snapshot:C\:\\Users\\kazsa11\\Zotero\\storage\\REJQHJWS\\2012.html:text/html;Full Text PDF:C\:\\Users\\kazsa11\\Zotero\\storage\\XYFSB5CG\\Morata-Dolz et al. - 2020 - Understanding Climate Impacts on Vegetation with G.pdf:application/pdf},
}
@misc{weichwald_causal_2020,
title = {Causal structure learning from time series: {Large} regression coefficients may predict causal links better in practice than small p-values},
shorttitle = {Causal structure learning from time series},
url = {http://arxiv.org/abs/2002.09573},
abstract = {In this article, we describe the algorithms for causal structure learning from time series data that won the Causality 4 Climate competition at the Conference on Neural Information Processing Systems 2019 (NeurIPS). We examine how our combination of established ideas achieves competitive performance on semi-realistic and realistic time series data exhibiting common challenges in real-world Earth sciences data. In particular, we discuss a) a rationale for leveraging linear methods to identify causal links in non-linear systems, b) a simulation-backed explanation as to why large regression coefficients may predict causal links better in practice than small p-values and thus why normalising the data may sometimes hinder causal structure learning. For benchmark usage, we detail the algorithms here and provide implementations at https://github.com/sweichwald/tidybench . We propose the presented competition-proven methods for baseline benchmark comparisons to guide the development of novel algorithms for structure learning from time series.},
urldate = {2024-04-16},
publisher = {arXiv},
author = {Weichwald, Sebastian and Jakobsen, Martin E. and Mogensen, Phillip B. and Petersen, Lasse and Thams, Nikolaj and Varando, Gherardo},
month = sep,
year = {2020},
note = {arXiv:2002.09573 [cs, stat]},
keywords = {Computer Science - Machine Learning, Statistics - Machine Learning, Statistics - Applications},
file = {arXiv.org Snapshot:C\:\\Users\\kazsa11\\Zotero\\storage\\HZD2RY8T\\2002.html:text/html;Full Text PDF:C\:\\Users\\kazsa11\\Zotero\\storage\\5UXG48AG\\Weichwald et al. - 2020 - Causal structure learning from time series Large .pdf:application/pdf},
}
@article{ramachandra_causal_nodate,
title = {Causal inference for climate change events from satellite image time series using computer vision and deep learning},
abstract = {We propose a method for causal inference using satellite image time series, in order to determine the treatment effects of interventions which impact climate change, such as deforestation. Simply put, the aim is to quantify the ‘before versus after’ effect of climate related human driven interventions, such as urbanization; as well as natural disasters, such as hurricanes and forest fires. As a concrete example, we focus on quantifying forest tree cover change/ deforestation due to human led causes. The proposed method involves the following steps. First, we uae computer vision and machine learning/deep learning techniques to detect and quantify forest tree coverage levels over time, at every time epoch. We then look at this time series to identify changepoints. Next, we estimate the expected (forest tree cover) values using a Bayesian structural causal model and projecting/forecasting the counterfactual. This is compared to the values actually observed post intervention, and the difference in the two values gives us the effect of the intervention (as compared to the non intervention scenario, i.e. what would have possibly happened without the intervention). As a specific use case, we analyze deforestation levels before and after the hyperinflation event (intervention) in Brazil (which ended in 1993-94), for the Amazon rainforest region, around Rondonia, Brazil. For this deforestation use case, using our causal inference framework can help causally attribute change/reduction in forest tree cover and increasing deforestation rates due to human activities at various points in time.},
language = {en},
author = {Ramachandra, Vikas},
file = {Ramachandra - Causal inference for climate change events from sa.pdf:C\:\\Users\\kazsa11\\Zotero\\storage\\MCBWGYMZ\\Ramachandra - Causal inference for climate change events from sa.pdf:application/pdf},
}
GitHub Events
Total
- Watch event: 5
Last Year
- Watch event: 5