Inference of Altruism and Intrinsic Rewards in Multi-Agent Systems

Human interactions are influenced by emotions, temperament, and affection, often conflicting with individuals' underlying preferences. Without explicit knowledge of those preferences, judging whether behaviour is appropriate becomes guesswork, leaving us highly prone to misinterpretation. Yet, such understanding is critical if autonomous agents are to collaborate effectively with humans. We frame this as a multi-agent inverse reinforcement learning problem and show that even a simple model, where agents weigh their own welfare against that of others, can cover complex social behaviours. Using novel Bayesian techniques, we find that intrinsic rewards and altruistic tendencies can be reliably identified by placing agents in varied groups. Crucially, this disentanglement of intrinsic motivation from altruism enables the synthesis of new behaviours aligned with any desired level of altruism, even when demonstrations are drawn from restricted behaviour profiles.