Reinfocement Learning Classes repository

Distributed Heuristic Multi-Agent Path Finding with Communication - ICRA 2021

强化学习中文教程（蘑菇书🍄），在线阅读地址：https://datawhalechina.github.io/easy-rl/

RL code accompanying our submission title "Adaptive Routing with Guaranteed Delay Bounds using Safe Reinforcement Learning"

Code accompanying our submission to RTSS 2019 titled " Adaptive Routing with Guaranteed Delay Bounds using Safe Reinforcement Learning"

Plan and execute routes in an urban environment given a destination location avoiding obstacles using Voronoi graphs and Probabilistic Road Maps (PRM). Build realistic paths by minimizing waypoints…

Distributed (and single-agent) reinforcement learning code for simulated multi-robot distributed assembly.

ROS package for dynamic obstacle avoidance for ground robots trained with deep RL

《李宏毅深度学习教程》（李宏毅老师推荐👍，苹果书🍎），PDF下载地址：https://github.com/datawhalechina/leedl-tutorial/releases

deeplearning.ai（吴恩达老师的深度学习课程笔记及资源）

This repository contains my personal notes and summaries on DeepLearning.ai specialization courses. I've enjoyed every little bit of the course hope you enjoy my notes too.

A collection of useful environments for testing Reinforcement Learning algorithms. Designed (mostly) with discrete, graph-based methods in mind.

This repository contains an attempt at using Graph Attention based Reinforcement Learning for graphical state space. The code also provides a visualisation for the attention weights

Optimal Action Space Search (OASS) is an algorithm for path planning problems on directed acyclic graphs (DAG) based on reinforcement learning (RL) theory.

Reinforcement learning implementations/experiments on graph-based/relational environments.

TorchGRL is the source code for our paper Graph Convolution-Based Deep Reinforcement Learning for Multi-Agent Decision-Making in Mixed Traffic Environments.

The code for AAMAS2022 《GCS: Graph-based Coordination Strategy for Multi-Agent Reinforcement Learning》

Graph-based Reinforcement Learning

Implementation of a Voronoi Diagram for Roadmap building and D* Lite for path planning and replanning for safe robot navigation in dynamic environnments.

An algorithm that uses Q-learning to find the optimal path to evacuate a building.

Official code for "DHRL: A Graph-Based Approach for Long-Horizon and Sparse Hierarchical Reinforcement Learning" (NeurIPS 2022 Oral)

We implemented a Q-learning algorithm that aims to find randomly selected erroneous nodes in a strongly connected directed graph. The models are also randomized based on the user-defined parameters…

A Q-Learning agent which finds the optimum escape route.

PRIMAL: Pathfinding via Reinforcement and Imitation Multi-Agent Learning -- Distributed RL/IL code for Multi-Agent Path Finding (MAPF)