The Emergence of Individuality

TL;DR

This paper introduces a novel method called EOI that promotes individuality in multi-agent reinforcement learning by using a probabilistic classifier and intrinsic rewards, leading to improved cooperation.

Contribution

The paper proposes a simple, efficient approach for the emergence of individuality in MARL using a probabilistic classifier and regularizers, enhancing agent discriminability and cooperation.

Findings

EOI outperforms existing methods in cooperative scenarios.

The intrinsic reward based on classifier prediction effectively encourages individuality.

Regularizers improve the discriminability of the classifier.

Abstract

Individuality is essential in human society, which induces the division of labor and thus improves the efficiency and productivity. Similarly, it should also be the key to multi-agent cooperation. Inspired by that individuality is of being an individual separate from others, we propose a simple yet efficient method for the emergence of individuality (EOI) in multi-agent reinforcement learning (MARL). EOI learns a probabilistic classifier that predicts a probability distribution over agents given their observation and gives each agent an intrinsic reward of being correctly predicted by the classifier. The intrinsic reward encourages the agents to visit their own familiar observations, and learning the classifier by such observations makes the intrinsic reward signals stronger and the agents more identifiable. To further enhance the intrinsic reward and promote the emergence of individuality, two regularizers are proposed to increase the discriminability of the classifier. We implement EOI on top of popular MARL algorithms. Empirically, we show that EOI significantly outperforms existing methods in a variety of multi-agent cooperative scenarios.