An Efficient Open World Environment for Multi-Agent Social Learning

TL;DR

This paper introduces a new open-ended multi-agent environment designed to facilitate research on social learning, cooperation, and competition among AI agents with complex goals, aiming to advance AI deployment in real-world scenarios.

Contribution

The work presents a novel environment for multi-agent social learning that supports complex, independent goals and emergent cooperation, enabling new research in socially intelligent AI.

Findings

Agents benefit from social learning with experts.

Emergent collaborative tool use observed among agents.

Cooperation and competition improve agent performance.

Abstract

Many challenges remain before AI agents can be deployed in real-world environments. However, one virtue of such environments is that they are inherently multi-agent and contain human experts. Using advanced social intelligence in such an environment can help an AI agent learn adaptive skills and behaviors that a known expert exhibits. While social intelligence could accelerate training, it is currently difficult to study due to the lack of open-ended multi-agent environments. In this work, we present an environment in which multiple self-interested agents can pursue complex and independent goals, reflective of real world challenges. This environment will enable research into the development of socially intelligent AI agents in open-ended multi-agent settings, where agents may be implicitly incentivized to cooperate to defeat common enemies, build and share tools, and achieve long horizon goals. In this work, we investigate the impact on agent performance due to social learning in the presence of experts and implicit cooperation such as emergent collaborative tool use, and whether agents can benefit from either cooperation or competition in this environment.