RL starter files in order to immediately train, visualize and evaluate an agent without writing any line of code

Multi-hop knowledge graph reasoning learned via policy gradient with reward shaping and action dropout

Inference-time scaling for LLMs-as-a-judge.

Recurrent and multi-process PyTorch implementation of deep reinforcement Actor-Critic algorithms A2C and PPO

Robotics research demonstrating reliability and robustness in the real world (continuously updated)

This repo implements our paper, "Learning to Search Feasible and Infeasible Regions of Routing Problems with Flexible Neural k-Opt", which has been accepted at NeurIPS 2023.

Implementation of Humanoid Standing Up, from the paper "Learning Humanoid Standing-up Control across Diverse Postures" out of Shanghai, in Pytorch

TraderNet-CRv2 - Combining Deep Reinforcement Learning with Technical Analysis and Trend Monitoring on Cryptocurrency Markets

Guide Your Agent with Adaptive Multimodal Rewards (NeurIPS 2023 Accepted)

Dota 2 bot that is trained by Deep RL with expert demonstrations

Code for NeurIPS 2022 paper Exploiting Reward Shifting in Value-Based Deep RL

Bayesian Reward Shaping Framework for Deep Reinforcement Learning

Code for "DrS: Learning Reusable Dense Rewards for Multi-Stage Tasks"

3D gym environments to train RL agents to play the Slime Volleyball game in 3 dimensions using Webots as simulator.

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.

Subtask-Aware Visual Reward Learning from Segmented Demonstrations (ICLR 2025 accepted)

Benchmarks for risk-aware reward shaping of autonomous driving

⚔️ Playing Jedi Academy with Deep Curiosity ⚔️

Set of experiments of using weights of a previously trained network as prior knowledge for a more complicated one and reward providing.