Too many cooks: Coordinating multi-agent collaboration through inverse planning
- Sarah Wu, Brain and Cognitive Sciences, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States
- Rose E Wang, MIT, Cambridge, Massachusetts, United States
- James Evans, University of Chicago, Chicago, Illinois, United States
- Josh Tenenbaum, Brain and Cognitive Sciences, MIT, Cambridge, Massachusetts, United States
- David Parkes, Harvard University, Cambridge, Massachusetts, United States
- Max Kleiman-Weiner, Brain and Cognitive Sciences, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States
AbstractCollaboration requires agents to coordinate their behavior on the fly, sometimes cooperating to solve a single task together and other times dividing it up into sub-tasks to work on in parallel. Underlying the human ability to collaborate is theory-of-mind, the ability to infer the hidden mental states that drive others to act. Here, we develop Bayesian Delegation, a decentralized multi-agent learning mechanism with these abilities. Bayesian Delegation enables agents to rapidly infer the hidden intentions of others by inverse planning. These inferences enable agents to flexibly decide in the absence of communication when to cooperate on the same sub-task and when to work on different sub-tasks in parallel. We test this model in a suite of multi-agent Markov decision processes inspired by cooking problems. To succeed, agents must coordinate both their high-level plans (e.g., what sub-task they should work on) and their low-level actions (e.g., avoiding collisions). Bayesian Delegation bridges these two levels and rapidly aligns agents' beliefs about who should work on what. Finally, we tested Bayesian Delegation in a behavioral experiment where participants made sub-task inferences from sparse observations of cooperative behavior. Bayesian Delegation outperformed heuristic models and was closely aligned with human judgments.