Benchmarking Robustness and Generalization in Multi-Agent Systems

A Case Study on Neural MMO

Journal Article (2023)
Author(s)

Yangkun Chen (Parametrix.ai, Tsinghua University)

Chenghui Yu (Parametrix.ai, Tsinghua University)

Hengman Zhu (Parametrix.ai)

Shuai Liu (Bilibili)

Yibing Zhang (Chengdu Goldwin Electronics Technology)

Joseph Suarez (Massachusetts Institute of Technology)

Liang Zhao (International Digital Economy Academy)

Jinke He (TU Delft - Interactive Intelligence)

Jiaxin Chen (Parametrix.ai)

undefined More Authors (External organisation)

Research Group
Interactive Intelligence
More Info
expand_more
Publication Year
2023
Language
English
Research Group
Interactive Intelligence
Bibliographical Note
Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.
Journal title
Proceedings of the International Joint Conference on Autonomous Agents and Multiagent Systems, AAMAS
Volume number
2023-May
Pages (from-to)
2490-2492
Event
22nd International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2023 (2023-05-29 - 2023-06-02), London, United Kingdom
Downloads counter
269
Collections
Institutional Repository
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

We present the results of the second Neural MMO challenge, hosted at IJCAI 2022, which received 1600+ submissions. This competition targets robustness and generalization in multi-agent systems: participants train teams of agents to complete a multi-task objective against opponents not seen during training. We summarize the competition design and results and suggest that, considering our work as a case study, competitions are an effective approach to solving hard problems and establishing a solid benchmark for algorithms. We will open-source our benchmark including the environment wrapper, baselines, a visualization tool, and selected policies for further research.

Files

3545946.3598978.pdf
(pdf | 3.42 Mb)
- Embargo expired in 27-11-2023
License info not available