reward-shaping
Implementation of Humanoid Standing Up, from the paper "Learning Humanoid Standing-up Control across Diverse Postures" out of Shanghai, in Pytorch
Inference-time scaling for LLMs-as-a-judge.
A gymnasium-compatible framework to create reinforcement learning (RL) environment for solving the optimal power flow (OPF) problem. Contains five OPF benchmark environments for comparable research.
Catch reward traps before training. Static analysis for RL reward functions.
Reward shaping library
