The impact of behavioral diversity in multi-agent reinforcement learning
This addresses the challenge of collective problem-solving in AI for applications like climate change and global peace, though it appears incremental by applying existing diversity paradigms to multi-agent reinforcement learning.
The paper tackles the problem of homogeneous agent strategies in multi-agent reinforcement learning by studying behavioral diversity, showing that controlling diversity yields non-trivial benefits such as improved team outcomes, better cooperation in sparse reward settings, and enhanced skill retention against disruptions.
Many of the world's most pressing issues, such as climate change and global peace, require complex collective problem-solving skills. Recent studies indicate that diversity in individuals' behaviors is key to developing such skills and increasing collective performance. Yet behavioral diversity in collective artificial learning is understudied, with today's machine learning paradigms commonly favoring homogeneous agent strategies over heterogeneous ones, mainly due to computational considerations. In this work, we employ diversity measurement and control paradigms to study the impact of behavioral heterogeneity in several facets of multi-agent reinforcement learning. Through experiments in team play and other cooperative tasks, we show the emergence of unbiased behavioral roles that improve team outcomes; how behavioral diversity synergizes with morphological diversity; how diverse agents are more effective at finding cooperative solutions in sparse reward settings; and how behaviorally heterogeneous teams learn and retain latent skills to overcome repeated disruptions. Overall, our results indicate that, by controlling diversity, we can obtain non-trivial benefits over homogeneous training paradigms, demonstrating that diversity is a fundamental component of collective artificial learning, an insight thus far overlooked.