2015/06/03 by Stefano V. Albrecht, Subramanian Ramamoorthy, Albrecht, Stefano V. +1 · 3 citations
Computer Science · Decision Sciences · #Artificial Intelligence (cs.AI) #Auction Theory and Applications #Computer Science and Game Theory (cs.GT) #FOS: Computer and information sciences #Game Theory and Applications #Multiagent Systems (cs.MA) #Reinforcement Learning in Robotics
paper · pdf · doi:10.48550/arxiv.1506.01170
openalex publication_date 2015/06/03 · openalex created_date 2022/09/02 · openalex updated_date 2026/07/28
The ad hoc coordination problem is to design an autonomous agent which is\nable to achieve optimal flexibility and efficiency in a multiagent system with\nno mechanisms for prior coordination. We conceptualise this problem formally\nusing a game-theoretic model, called the stochastic Bayesian game, in which the\nbehaviour of a player is determined by its private information, or type. Based\non this model, we derive a solution, called Harsanyi-Bellman Ad Hoc\nCoordination (HBA), which utilises the concept of Bayesian Nash equilibrium in\na planning procedure to find optimal actions in the sense of Bellman optimal\ncontrol. We evaluate HBA in a multiagent logistics domain called level-based\nforaging, showing that it achieves higher flexibility and efficiency than\nseveral alternative algorithms. We also report on a human-machine experiment at\na public science exhibition in which the human participants played repeated\nPrisoner's Dilemma and Rock-Paper-Scissors against HBA and alternative\nalgorithms, showing that HBA achieves equal efficiency and a significantly\nhigher welfare and winning rate.\n