roderickli / marl-papers-with-code

Multi-Agent Reinforcement Learning (MARL) papers with code

MARL Papers with Code

This is a collection of Multi-Agent Reinforcement Learning (MARL) papers with code. I have selected some relatively important papers with open source code and categorized them by time and method.

For MARL papers and MARL resources, please refer to Multi Agent Reinforcement Learning papers and MARL Resources Collection.

I will continually update this repository and I welcome suggestions. (missing important papers, missing categories, invalid links, etc.) This is only a first draft so far and I'll add more resources in the next few months.

This repository is not for commercial purposes.

My email: [email protected]

Overview

• Classic Papers
• Other Papers
• TODO

Classic Papers

Category	Paper	Code
Independent Learning	IQL：Multi-Agent Reinforcement Learning: Independent vs. Cooperative Agents	https://github.com/oxwhirl/pymarl
Value Decomposition	VDN：Value-Decomposition Networks For Cooperative Multi-Agent Learning	https://github.com/oxwhirl/pymarl
Value Decomposition	QMIX: Monotonic Value Function Factorisation for Deep Multi-Agent Reinforcement Learning	https://github.com/oxwhirl/pymarl
Value Decomposition	QTRAN: Learning to Factorize with Transformation for Cooperative Multi-Agent Reinforcement Learning	https://github.com/oxwhirl/pymarl
Policy Gradient	COMA：Counterfactual Multi-Agent Policy Gradients	https://github.com/oxwhirl/pymarl
Policy Gradient	MADDPG：Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments	https://github.com/openai/maddpg
Communication	BiCNet：Multiagent Bidirectionally-Coordinated Nets: Emergence of Human-level Coordination in Learning to Play StarCraft Combat Games	https://github.com/Coac/CommNet-BiCnet
Communication	CommNet：Learning Multiagent Communication with Backpropagation	https://github.com/facebookarchive/CommNet
Communication	IC3Net：Learning when to Communicate at Scale in Multiagent Cooperative and Competitive Tasks	https://github.com/IC3Net/IC3Net
Communication	RIAL/RIDL：Learning to Communicate with Deep Multi-Agent Reinforcement Learning	https://github.com/iassael/learning-to-communicate
Exploration	MAVEN：Multi-Agent Variational Exploration	https://github.com/starry-sky6688/StarCraft

Other Papers

Category	Paper	Code
Graph Neural Network	Multi-Agent Game Abstraction via Graph Attention Neural Network	https://github.com/starry-sky6688/StarCraft
Curriculum Learning	From Few to More: Large-Scale Dynamic Multiagent Curriculum Learning	https://github.com/starry-sky6688/StarCraft
Curriculum Learning	EPC：Evolutionary Population Curriculum for Scaling Multi-Agent Reinforcement Learning	https://github.com/qian18long/epciclr2020
Curriculum Learning/Emergent	Emergent Tool Use From Multi-Agent Autocurricula	https://github.com/openai/multi-agent-emergence-environments
Curriculum Learning	Cooperative Multi-agent Control using deep reinforcement learning	https://github.com/sisl/MADRL
Role	ROMA: Multi-Agent Reinforcement Learning with Emergent Roles	https://github.com/TonghanWang/ROMA
Role	RODE: Learning Roles to Decompose Multi-Agent Tasks	https://github.com/TonghanWang/RODE
Opponent Modeling	Opponent Modeling in Deep Reinforcement Learning	https://github.com/hhexiy/opponent
Selfish Agent	Mind-aware Multi-agent Management Reinforcement Learning	https://github.com/facebookresearch/M3RL
Communication	Emergence of grounded compositional language in multi-agent populations	https://github.com/bkgoksel/emergent-language
Communication	Fully decentralized multi-agent reinforcement learning with networked agents	https://github.com/cts198859/deeprl_network
Policy Gradient	DOP: Off-Policy Multi-Agent Decomposed Policy Gradients	https://github.com/TonghanWang/DOP
Policy Gradient	Actor-Attention-Critic for Multi-Agent Reinforcement Learning	https://github.com/shariqiqbal2810/MAAC
Environment	Emergent Complexity via Multi-Agent Competition	https://github.com/openai/multiagent-competition
Exploration	EITI/EDTI：Influence-Based Multi-Agent Exploration	https://github.com/TonghanWang/EITI-EDTI
Exploration	LIIR: Learning Individual Intrinsic Reward in Multi-Agent Reinforcement Learning	https://github.com/yalidu/liir
From Single-Agent to Multi-Agent	MAPPO：The Surprising Effectiveness of MAPPO in Cooperative, Multi-Agent Games	https://github.com/marlbenchmark/on-policy
Diversity	Q-DPP：Multi-Agent Determinantal Q-Learning	https://github.com/QDPP-GitHub/QDPP
Ad Hoc Teamwork	CollaQ：Multi-Agent Collaboration via Reward Attribution Decomposition	https://github.com/facebookresearch/CollaQ
Value Decomposition	NDQ: Learning Nearly Decomposable Value Functions Via Communication Minimization	https://github.com/TonghanWang/NDQ
Value Decomposition	QPLEX: Duplex Dueling Multi-Agent Q-Learning	https://github.com/wjh720/QPLEX
Self-Play	TLeague: A Framework for Competitive Self-Play based Distributed Multi-Agent Reinforcement Learning	https://github.com/tencent-ailab/TLeague
Others	UPDeT: Universal Multi-agent Reinforcement Learning via Policy Decoupling with Transformers	https://github.com/hhhusiyi-monash/UPDeT
Others	M3RL: Mind-aware Multi-agent Management Reinforcement Learning	https://github.com/facebookresearch/M3RL

TODO

• More papers will be added to Other Papers.