A Platform for Many-agent Reinforcement Learning

Reinforcement Learning / AI Bots in Card (Poker) Games - Blackjack, Leduc, Texas, DouDizhu, Mahjong, UNO.

Foundation is a flexible, modular, and composable framework to model socio-economic behaviors and dynamics with both agents and governments. This framework can be used in conjunction with reinforcement learning to learn optimal economic policies, as done by the AI Economist (https://www.einstein.ai/the-ai-economist).

PyTorch implementations of various Deep Reinforcement Learning (DRL) algorithms for both single agent and multi-agent.

Multi-Agent Resource Optimization (MARO) platform is an instance of Reinforcement Learning as a Service (RaaS) for real-world resource optimization. It can be applied to many important industrial domains, such as container inventory management in logistics, bike repositioning in transportation, virtual machine provisioning in data centers, and asset management in finance. Besides RL, it also supports other planning/decision mechanisms, such as Operations Research. MARO provides comprehensive support in data processing, simulator building, RL algorithm selection, and distributed training.

A collection of multi agent environments based on OpenAI gym.

Awesome Game AI materials of Multi-Agent Reinforcement Learning

The multi-agent version of TORCS for developing control algorithms for fully autonomous driving in the cluttered, multi-agent settings of everyday life.

Block-style programming environment for multi agent system, based on MIT Scratch.

Multi-Agent Robotics Simulator

TensorSwarm: A framework for reinforcement learning of robot swarms.

API to run VirtualHome, a Multi-Agent Household Simulator

BARK - a tool for Behavior benchmARKing.

PyBullet Gym environments for single and multi-agent reinforcement learning of quadcopter control

Mobile Agent that moves on a platform distributed on many machines and recover the mac address of each machine during the deplacement

A Universal Deep Reinforcement Learning Framework

MultiAgent Development Kit - version 5

Multi & Single Agent Reinforcement Learning for Traffic Signal Control Problem

Algorithm for prioritized multi-agent path finding (MAPF) in grid-worlds. Moves into arbitrary directions are allowed (each agent is allowed to follow any-angle path on the grid). Timeline is continuous, i.e. action durations are not explicitly discretized into timesteps. Different agents' size and moving speed are supported. Planning is carried out in (x, y, \theta) configuration space, i.e. agents' orientation are taken into account.

Lightweight multi-agent gridworld Gym environment

reproduce some RL or Multi-Agent models

This is an implementation of the paper "Coordinated Multi Agent Imitation Learning", or the Sloan version "Data-Driven Ghosting using Deep Imitation Learning" using Tensorflow

This repository contains the Matlab source codes (to use in Robotarium platform) of various rendezvous controllers for consensus control in a multi-agent / multi-robot system.

Continuous CBS - a modification of conflict based search algorithm, that allows to perform actions (move, wait) of arbitrary duration. Timeline is not discretized, i.e. is continuous.

Collection of Deep Reinforcement Learning Algorithms implemented in PyTorch.

Continuous-time multi-agent crowd simulation engine implemented in Python using Numba and Numpy for performance.

Negotiation Multi-Agent System (A negotiation library designed for situated negotiations within business-like simulations)

Multi-agent deep reinforcement learning research