A Predator-Prey-Grass multi-objective multi-agent gridworld environment implemented with Farama's Gymnasium, PettingZoo and MOMAland, featuring dynamic agent spawning and deletion, where agents have partial observability.

All my GitHub projects in one place.

A modular, primitive-first, python-first PyTorch library for Reinforcement Learning.

🦁 A research-friendly codebase for fast experimentation of multi-agent reinforcement learning in JAX

Extended, multi-agent and multi-objective (MaMoRL / MoMaRL) environments based on DeepMind's AI Safety Gridworlds. This is a suite of reinforcement learning environments illustrating various safety properties of intelligent agents. It is made compatible with OpenAI's Gym/Gymnasium and Farama Foundation PettingZoo.

Develop your agent for generals.io!

A collection of MARL benchmarks based on TorchRL

VMAS is a vectorized differentiable simulator designed for efficient Multi-Agent Reinforcement Learning benchmarking. It is comprised of a vectorized 2D physics engine written in PyTorch and a set of challenging multi-robot scenarios. Additional scenarios can be implemented through a simple and modular interface.

POGEMA stands for Partially-Observable Grid Environment for Multiple Agents. This is a grid-based environment that was specifically designed to be flexible, tunable and scalable. It can be tailored to a variety of PO-MAPF settings.

SigmaRL: A Sample-Efficient and Generalizable Multi-Agent Reinforcement Learning Framework for Motion Planning

This is a framework for the research on multi-agent reinforcement learning and the implementation of the experiments in the paper titled by ''Shapley Q-value: A Local Reward Approach to Solve Global Reward Games''.

A tool for aggregating and plotting MARL experiment data.

Pytorch implementation of MAPPO algo in CleanRL style

This repository considers the implementation of the paper "FoX: Formation-aware exploration in multi-agent reinforcement learning" which has been accepted to AAAI 2024

Multi-Agent Deep Reinforcement Learning for Cooperative and Competitive Autonomous Vehicles

An introduction of our NeurIPS 2024 paper: CORY, please star the origin repo for more details

This repository contains the code for Diversity Control (DiCo), a novel method to constrain behavioral diversity in multi-agent reinforcement learning.

Multi-Objective Multi Agent Reinforcement Learning (MOMARL) experiments.

Multi-agent collaboration using Reinforcement Learning

PPO and PyMARL baseline for Pogema environment