https://github.com/aida-ugent/fipr
The KL-Divergence between a Graph Model and its Fair I-Projection as a Fairness Regularizer (ECML-PKDD 2021).
Science Score: 13.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
○CITATION.cff file
-
✓codemeta.json file
Found codemeta.json file -
○.zenodo.json file
-
○DOI references
-
○Academic publication links
-
○Academic email domains
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (11.4%) to scientific vocabulary
Keywords
Repository
The KL-Divergence between a Graph Model and its Fair I-Projection as a Fairness Regularizer (ECML-PKDD 2021).
Basic Info
Statistics
- Stars: 1
- Watchers: 3
- Forks: 0
- Open Issues: 0
- Releases: 0
Topics
Metadata Files
README.md
Fair I-Projection Regularizer
Experiment code for the paper "The KL-Divergence between a Graph Model and its Fair I-Projection as a Fairness Regularizer", published at ECML-PKDD 2021.
Running the code
Running main_all.py executes the pipeline as configured by config.py. NOTE: this will overwrite any results in the results/ folder.
Alternatively, run main_simple.py for a simple example usage.
Setup
1) Download the desired datasets. See the data/ folder for their links.
2) Install the required packages as indicated by requirements.txt.
Fair I-Projection as a regularizer in your projects
The expected interface of each (link) Predictor is documented in predictor.py.
The distance from a model to its I-projection can be computed using the FairnessLoss PyTorch module. A forward call on this module consists of two steps. First, the fair I-projection (code in fip.py) is fit to the given model values h and data points x. Second, the gradient of the fair I-projection's loss is computed with respect to the model values h.
Two fairness notions are implemented in fairness_notions.py: 'DP' for Demographic Parity and 'EO' for Equalised Opportunity.
Citation
If you found our work useful in your own project, please cite our paper:
@inproceedings{buyl2021kl,
author={Buyl, Maarten and De Bie, Tijl},
title={The KL-Divergence Between a Graph Model and its Fair I-Projection as a Fairness Regularizer},
booktitle={Machine Learning and Knowledge Discovery in Databases},
year={2021},
publisher={Springer International Publishing},
pages={351--366}
}
Maintenance
Further development may be done in the future and bugs will be fixed. If you have any questions or concerns, feel free to report it here or send an email to 'maarten.buyl@ugent.be'.
Owner
- Name: Ghent University Artificial Intelligence & Data Analytics Group
- Login: aida-ugent
- Kind: organization
- Email: tijl.debie@ugent.be
- Location: Ghent
- Website: aida.ugent.be
- Repositories: 36
- Profile: https://github.com/aida-ugent
GitHub Events
Total
Last Year
Issues and Pull Requests
Last synced: 11 months ago
All Time
- Total issues: 0
- Total pull requests: 0
- Average time to close issues: N/A
- Average time to close pull requests: N/A
- Total issue authors: 0
- Total pull request authors: 0
- Average comments per issue: 0
- Average comments per pull request: 0
- Merged pull requests: 0
- Bot issues: 0
- Bot pull requests: 0
Past Year
- Issues: 0
- Pull requests: 0
- Average time to close issues: N/A
- Average time to close pull requests: N/A
- Issue authors: 0
- Pull request authors: 0
- Average comments per issue: 0
- Average comments per pull request: 0
- Merged pull requests: 0
- Bot issues: 0
- Bot pull requests: 0
Top Authors
Issue Authors
Pull Request Authors
Top Labels
Issue Labels
Pull Request Labels
Dependencies
- networkx ==2.5
- numpy ==1.19.2
- pandas ==1.1.5
- scikit_learn ==0.24.1
- scipy ==1.5.2
- torch ==1.7.1
- tqdm ==4.54.1