CaltechAUTHORS
  A Caltech Library Service

Coordinated Multi-Agent Imitation Learning

Le, Hoang M. and Yue, Yisong and Carr, Peter and Lucey, Patrick (2017) Coordinated Multi-Agent Imitation Learning. Proceedings of Machine Learning Research, 70 . pp. 1995-2003. ISSN 1938-7228. https://resolver.caltech.edu/CaltechAUTHORS:20170829-143913716

[img] PDF - Published Version
See Usage Policy.

2081Kb
[img] PDF - Accepted Version
See Usage Policy.

2207Kb
[img] PDF (arXiv) - Accepted Version
See Usage Policy.

2239Kb

Use this Persistent URL to link to this item: https://resolver.caltech.edu/CaltechAUTHORS:20170829-143913716

Abstract

We study the problem of imitation learning from demonstrations of multiple coordinating agents. One key challenge in this setting is that learning a good model of coordination can be difficult, since coordination is often implicit in the demonstrations and must be inferred as a latent variable. We propose a joint approach that simultaneously learns a latent coordination model along with the individual policies. In particular, our method integrates unsupervised structure learning with conventional imitation learning. We illustrate the power of our approach on a difficult problem of learning multiple policies for fine-grained behavior modeling in team sports, where different players occupy different roles in the coordinated team strategy. We show that having a coordination model to infer the roles of players yields substantially improved imitation loss compared to conventional baselines.


Item Type:Article
Related URLs:
URLURL TypeDescription
http://proceedings.mlr.press/v70/le17a.htmlPublisherArticle
https://arxiv.org/abs/1703.03121arXivArticle
Additional Information:© 2017 the authors. This work was funded in part by NSF Awards #1564330 & #1637598, JPL PDF IAMS100224, a Bloomberg Data Science Research Grant, and a gift from Northrop Grumman.
Funders:
Funding AgencyGrant Number
NSFIIS-1564330
NSFCCF-1637598
JPLIAMS100224
Bloomberg Data ScienceUNSPECIFIED
Northrop GrummanUNSPECIFIED
Record Number:CaltechAUTHORS:20170829-143913716
Persistent URL:https://resolver.caltech.edu/CaltechAUTHORS:20170829-143913716
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:80920
Collection:CaltechAUTHORS
Deposited By: Tony Diaz
Deposited On:30 Aug 2017 17:12
Last Modified:03 Oct 2019 18:36

Repository Staff Only: item control page