Multi-Agent Interplay in a Competitive Survival Environment

TL;DR

This paper introduces a new competitive multi-agent environment with realistic physics and semantics, demonstrating emergent behaviors and strategies, advancing understanding of multi-agent interactions in complex settings.

Contribution

It presents an extensible environment for multi-agent interplay with realistic physics, and provides experimental insights into emergent strategies in competitive scenarios.

Findings

Emergent simple strategies observed in experiments

Environment features realistic physics and semantics

Directions for future improvements identified

Abstract

Solving hard-exploration environments in an important challenge in Reinforcement Learning. Several approaches have been proposed and studied, such as Intrinsic Motivation, co-evolution of agents and tasks, and multi-agent competition. In particular, the interplay between multiple agents has proven to be capable of generating human-relevant emergent behaviour that would be difficult or impossible to learn in single-agent settings. In this work, an extensible competitive environment for multi-agent interplay was developed, which features realistic physics and human-relevant semantics. Moreover, several experiments on different variants of this environment were performed, resulting in some simple emergent strategies and concrete directions for future improvement. The content presented here is part of the author's thesis "Multi-Agent Interplay in a Competitive Survival Environment" for the Master's Degree in Artificial Intelligence and Robotics at Sapienza University of Rome, 2022.