Bayesian Action Decoder for Deep Multi-Agent Reinforcement Learning

Abstract

When observing the actions of others, humans make inferences about why they acted as they did, and what this implies about the world; humans also use the fact that their actions will be interpreted in this manner, allowing them to act informatively and thereby communicate efficiently with others. Although learning algorithms have recently achieved superhuman performance in a number of two-player, zero-sum games, scalable multi-agent reinforcement learning algorithms that can discover effective strategies and conventions in complex, partially observable settings have proven elusive. We present the Bayesian action decoder (BAD), a new multi-agent learning method that uses an approximate Bayesian update to obtain a public belief that conditions on the actions taken by all agents in the environment. BAD introduces a new Markov decision process, the public belief MDP, in which the action space consists of all deterministic partial policies, and exploits the fact that an agent acting only on this public belief state can still learn to use its private information if the action space is augmented to be over all partial policies mapping private information into environment actions. The Bayesian update is closely related to the theory of mind reasoning that humans carry out when observing others' actions. We first validate BAD on a proof-of-principle two-step matrix game, where it outperforms policy gradient methods; we then evaluate BAD on the challenging, cooperative partial-information card game Hanabi, where, in the two-player setting, it surpasses all previously published learning and hand-coded approaches, establishing a new state of the art.

Cited by in corpus (24)

• Deep Reinforcement Learning for Multi-Agent Systems: A Review of Challenges, Solutions and Applications
• The Hanabi Challenge: A New Frontier for AI Research
• OpenSpiel: A Framework for Reinforcement Learning in Games
• Combining Deep Reinforcement Learning and Search for Imperfect-Information Games
• Simplified Action Decoder for Deep Multi-Agent Reinforcement Learning
• On the Critical Role of Conventions in Adaptive Human-AI Collaboration
• Emergent Multi-Agent Communication in the Deep Learning Era
• Evaluation of Human-AI Teams for Learned and Rule-Based Agents in Hanabi
• Promoting Coordination through Policy Regularization in Multi-Agent Deep Reinforcement Learning
• Re-determinizing Information Set Monte Carlo Tree Search in Hanabi
• Efficient Deviation Types and Learning for Hindsight Rationality in Extensive-Form Games
• Learning to Communicate Implicitly By Actions
• Neural Recursive Belief States in Multi-Agent Reinforcement Learning
• Continuous Coordination As a Realistic Scenario for Lifelong Learning
• Scalable Online Planning via Reinforcement Learning Fine-Tuning
• Incorporating Pragmatic Reasoning Communication into Emergent Language
• Evaluating the Rainbow DQN Agent in Hanabi with Unseen Partners
• Optimal communication and control strategies in a multi-agent MDP problem
• Learned Belief Search: Efficiently Improving Policies in Partially Observable Settings
• Exploring Zero-Shot Emergent Communication in Embodied Multi-Agent Populations
• Off-Belief Learning
• Common Information based Approximate State Representations in Multi-Agent Reinforcement Learning
• Distributed Policy Gradient with Variance Reduction in Multi-Agent Reinforcement Learning
• Adversarial Attacks in Cooperative AI