Intrinsic Motivation

Adversarial Surprise creates a competitive game between an Expore policy and a Control policy, which fight to maximize and minimize the amount of entropy an RL agent experiences. We show both theoretically and empirically that this technique fully explores the state space of partially-observed, stochastic environments.

Joint attention is a critical component of human social cognition. In this paper, we ask whether a mechanism based on shared visual attention can be useful for improving multi-agent coordination and social learning.

Social influence is a unified mechanism for achieving coordination and communication in Multi-Agent Reinforcement Learning, through rewarding agents for having causal influence over other agents’ actions, thus increasing mutual information between agents’ actions. Optimizing for influence leads to agents learning emergent communication protocols. Unlike prior work, influence can be computed in a fully decentralized manner.