Official Implementations of "Provably Efficient RL under Episode-Wise Safety in Constrained MDPs with Linear Function Approximation"

• See the implementation in Experiment.ipynb.

Requirements

uv sync

Run Experiments and Plot the results

Run all the cells in Experiment.ipynb. The results will be saved as "results-random.pdf" for the tabular env, "results-streaming.pdf" for the streaming env, and "results-linear.pdf" for the linear env.