Developing, Evaluating and Scaling Learning Agents in Multi-Agent Environments
Multiagent Systems
2022-09-23 v1 Artificial Intelligence
Abstract
The Game Theory & Multi-Agent team at DeepMind studies several aspects of multi-agent learning ranging from computing approximations to fundamental concepts in game theory to simulating social dilemmas in rich spatial environments and training 3-d humanoids in difficult team coordination tasks. A signature aim of our group is to use the resources and expertise made available to us at DeepMind in deep reinforcement learning to explore multi-agent systems in complex environments and use these benchmarks to advance our understanding. Here, we summarise the recent work of our team and present a taxonomy that we feel highlights many important open challenges in multi-agent research.
Cite
@article{arxiv.2209.10958,
title = {Developing, Evaluating and Scaling Learning Agents in Multi-Agent Environments},
author = {Ian Gemp and Thomas Anthony and Yoram Bachrach and Avishkar Bhoopchand and Kalesha Bullard and Jerome Connor and Vibhavari Dasagi and Bart De Vylder and Edgar Duenez-Guzman and Romuald Elie and Richard Everett and Daniel Hennes and Edward Hughes and Mina Khan and Marc Lanctot and Kate Larson and Guy Lever and Siqi Liu and Luke Marris and Kevin R. McKee and Paul Muller and Julien Perolat and Florian Strub and Andrea Tacchetti and Eugene Tarassov and Zhe Wang and Karl Tuyls},
journal= {arXiv preprint arXiv:2209.10958},
year = {2022}
}
Comments
Published in AI Communications 2022