AI Paper Atlas

Research map / Multi-agent systems and games / Games and equilibria

A Unified Game-Theoretic Approach to Multiagent Reinforcement Learning

Marc Lanctot, Vinicius Zambaldi, Audrunas Gruslys, Angeliki Lazaridou, Karl Tuyls, Julien Perolat, David Silver, Thore Graepel

NeurIPS 2017 · Multi-agent systems and games › Games and equilibria · approach: Learning in Markov games

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use DCH rather than InRLMultiagent policies · Laser Tag small2average proportional loss (R_-): 0.342 -> 0.055Improved
Use DCH rather than InRLMultiagent policies · Laser Tag small3average proportional loss (R_-): 0.625 -> 0.082Improved
Use DCH rather than InRLMultiagent policies · Laser Tag small4average proportional loss (R_-): 0.717 -> 0.150Improved
Use DCH rather than InRLMultiagent policies · Gathering fieldaverage proportional loss (R_-): 0.003 -> 0.007Hurt
Use DCH rather than InRLMultiagent policies · Pathfind mergeaverage proportional loss (R_-): 0.022 -> < 0Improved
Use the full mixed meta-strategy rather than only the highest-level policy (ablation)DCH · Laser Tag small2, small3, and small4average proportional loss (R_-): 0.147, 0.27, 0.118 for highest-level policies; full-strategy values 0.055, 0.082, 0.150Improved
Use level 5 rather than level 10 (ablation)DCH · Laser Tag small4average proportional loss (R_-): 0.156 vs. 0.150No clear effect
Use level 3 rather than level 10 (ablation)DCH · Laser Tag small4average proportional loss (R_-): 0.246 vs. 0.150Hurt
Use decoupled PRD rather than decoupled regret-matching or Exp3 (ablation)DCH · Two-player Leduc PokerNashConv MAUCImproved
Increase the DCH level (ablation)DCH · Two-player Leduc PokerNashConvImproved
Use PSRO/DCH rather than NFSPLeduc Poker · Leduc PokerNashConvMixed
Use PSRO/DCH rather than NFSPLeduc Poker · Leduc Poker against fixed playersperformance against fixed playersImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy