Safe RL Team's repositories
viper-verifiable-rl-impl
Implementation of the VIPER algorithm introduced in "Verifiable Reinforcement Learning via Policy Extraction" by Bastani et al.
topics-in-RL
A compilation of recent machine learning papers focused on safe reinforcement learning
curriculum-learning-poster
Poster about Curriculum Induction for Safe Reinforcement Learning
Uncertainty-Based-Offline-RL-with-Diversified-Q-Ensemble
Blog post about Uncertainty-Based Offline Reinforcement Learning with Diversified Q-Ensemble
Adaptive-Reward-Penalty-in-Safe-Reinforcement-Learning
In this work we have implemented RCPO into PPO and recreated the results from the original paper. This Blog post summarises our work and elaborates on our ideas and findings.
CARL-params
Code: Caution Parameters in "Cautious Adaptation for Reinforcement Learning in Safety-Critical Settings"
curriculum-learning
Blog Post about Curriculum Induction for Safe Reinforcement Learning
presentations
slides presenting state-of-art papers on safe reinforcement learning
viper-verifiable-reinforcement-learning
The blog post accompanying the implementation of the paper "Viper: Verifiable Reinforcement Learning via Policy Extraction" by Bastani et al.
adversarial-policies-pytorch-blog
Blog post for our implementation of the paper "Adversarial Policies: Attacking Deep Reinforcement Learning"
barrier-certificates
Code for Barrier Certificates Blog: https://safe-rl-team.github.io/barrier-certificates/
PID
Blog post about Responsive Safety in Reinforcement Learning by PID Lagrangian Methods
lambda-bo-blog
Lambda Bayesian optimization blog
post--example
Example Distill article repository—clone, rename, start writing!
Uncertainty-Based-Offline-RL-with-Diversified-Q-Ensemble-Implementation
This is a reimplementation of the EDAC algorithm in PyTorch. It was created as part of an University project and used for a blog post: https://github.com/Safe-RL-Team/Uncertainty-Based-Offline-RL-with-Diversified-Q-Ensemble