High-quality single file implementation of Deep Reinforcement Learning algorithms with research-friendly features (PPO, DQN, C51, DDPG, TD3, SAC, PPG)

Characteristic Value Iteration (CVI): frequency-domain reinforcement learning using characteristic functions on tabular MDPs

[Adage Lab version] Official Code for "Relative Entropy Pathwise Policy Optimization"

Official code for the paper "Calibrated Value-Aware Model Learning with Stochastic Environment Models

[ICML'25] PANDAS: Improving Many-shot Jailbreaking via Positive Affirmation, Negative Demonstration, and Adaptive Sampling

Official Code for the paper: Deflated Dynamics Value Iteratioon

Official package for the MDOT-TNT algorithm for discrete optimal transport.

Code to reproduce the results in the NeurIPS 2023 paper Distributional Model Equivalence for Risk-Sensitive Reinforcement Learning.

[ECCV'24] Improving Adversarial Transferability via Model Alignment

An application of ideas from control theory to hopefully accelerate the dynamics of TD learning.

Experiments for policy-aware model learning