ClosedLoopReachability
Reachability analysis for closed-loop control systems in Julia
astroaugmentations
A package with various custom augmentations implemented which are specifically designed around astronomical data.
stable-gym
This package contains several gymnasium environments with positive definite cost functions, designed for compatibility with stable RL agents.
thetis
Service to examine data processing pipelines (e.g., machine learning or deep learning pipelines) for uncertainty consistency (calibration), fairness, and other safety-relevant aspects.
https://github.com/bethgelab/adversarial-vision-challenge
NIPS Adversarial Vision Challenge
https://github.com/aeturrell/specification_curve
Specification Curve is a Python package that performs specification curve analysis: exploring how a coefficient varies under multiple different specifications of a statistical model.
https://github.com/holistic-ai/holisticai
This is an open-source tool to assess and improve the trustworthiness of AI systems.
stable-learning-control
A framework for training theoretically stable (and robust) Reinforcement Learning control algorithms.
https://github.com/alexstead/rfrontier
Stata package for robust stochastic frontier analysis
https://github.com/amazon-science/recode
Releasing code for "ReCode: Robustness Evaluation of Code Generation Models"
https://github.com/amazon-science/factual-confidence-of-llms
Code for paper "Factual Confidence of LLMs: on Reliability and Robustness of Current Estimators"
hyphi-gym
A Gymnasium benchmark suite for evaluating the robustness and multi-task performance of reinforcement learning algorithms in various discrete and continuous environments.
adversarial-nonparametrics
Robustness for Non-Parametric Classification: A Generic Attack and Defense
understanding-clip-ood
Official code for the paper: "When and How Does CLIP Enable Domain and Compositional Generalization?" (ICML 2025 Spotlight)
safe-control-gym
PyBullet CartPole and Quadrotor environments—with CasADi symbolic a priori dynamics—for learning-based control and RL
attention-meets-perturbation
📝 Official Implementation of "Attention Meets Perturbation: Robust and Interpretable Attention with Adversarial Training"
https://github.com/bethgelab/model-vs-human
Benchmark your model on out-of-distribution datasets with carefully collected human comparison data (NeurIPS 2021 Oral)
seer
A GAN (generative adversarial network) for projecting synthetic building performance profiles.
cvpr22w_robustnessthroughthelens
Official repository of our submission "Adversarial Robustness through the Lens of Convolutional Filters" for the CVPR2022 Workshop "The Art of Robustness: Devil and Angel in Adversarial Machine Learning Workshop"