Generative Multi-Agent Behavioral Cloning
Abstract
We propose and study the problem of generative multi-agent behavioral cloning, where the goal is to learn a generative, i.e., non-deterministic, multi-agent policy from pre-collected demonstration data. Building upon advances in deep generative models, we present a hierarchical policy framework that can tractably learn complex mappings from input states to distributions over multi-agent action spaces by introducing a hierarchy with macro-intent variables that encode long-term intent. In addition to synthetic settings, we show how to instantiate our framework to effectively model complex interactions between basketball players and generate realistic multi-agent trajectories of basketball gameplay over long time periods. We validate our approach using both quantitative and qualitative evaluations, including a user study comparison conducted with professional sports analysts.
Additional Information
This research is supported in part by NSF #1564330, NSF #1637598, and gifts from Bloomberg, Activision/Blizzard and Northrop Grumman. Dataset was provided by STATS: https://www.stats.com/data-science/.Attached Files
Submitted - 1803.07612.pdf
Files
Name | Size | Download all |
---|---|---|
md5:985805d468a461a5f24ea9f664732f6f
|
1.9 MB | Preview Download |
Additional details
- Eprint ID
- 92669
- Resolver ID
- CaltechAUTHORS:20190205-111434225
- IIS-1564330
- NSF
- CCF-1637598
- NSF
- Bloomberg Data Science
- Activision/Blizzard
- Northrop Grumman
- Created
-
2019-02-05Created from EPrint's datestamp field
- Updated
-
2023-06-02Created from EPrint's last_modified field