Visualizing the Evolution of Multi-agent Game-playing Behaviors

In the proceedings of EuroVis - Poster, 2022

Abstract:

Analyzing the training evolution of AI agents in a multi-agent environment helps to understand changes in learned behaviors, as well as the sequence in which they are learned. We train an existing Pommerman team from scratch and, at regular intervals, let it battle against another top-performing team. We define thirteen game-specific behaviors and compute their occurrences in 600 matches. To investigate the evolution of these behaviors, we propose a visualization approach and showcase its usefulness in an application example.

PDF Demo Poster DOI

To cite: Shivam Agarwal, Shahid Latif, Aristide Rothweiler, Fabian Beck, "Visualizing the Evolution of Multi-agent Game-playing Behaviors" In the proceedings of EuroVis - Poster, 2022. 10.2312/evp.20221111

BibTeX:

@inproceedings{Agarwal2022Visualizing,
author = {Agarwal, Shivam and Latif, Shahid and Rothweiler, Aristide and Beck, Fabian},
booktitle = {{EuroVis} - Poster},
title = {Visualizing the Evolution of Multi-agent Game-playing Behaviors},
abstract = {Analyzing the training evolution of AI agents in a multi-agent environment helps to understand changes in learned behaviors, as well as the sequence in which they are learned. We train an existing Pommerman team from scratch and, at regular intervals, let it battle against another top-performing team. We define thirteen game-specific behaviors and compute their occurrences in 600 matches. To investigate the evolution of these behaviors, we propose a visualization approach and showcase its usefulness in an application example.},
year = {2022},
ISBN = {978-3-03868-185-4},
DOI = {10.2312/evp.20221111}
}