Access Restriction

Source CiteSeerX
Content type Text
File Format PDF
Subject Domain (in DDC) Computer science, information & general works ♦ Data processing & computer science
Subject Keyword Model-based Reinforcement Learning ♦ Partially Observable Game ♦ Appropriate Strategy ♦ Card Game Heart ♦ Plausible Number ♦ Markov Chain Monte Carlo ♦ Heavy Integration ♦ Computer Simulation Result ♦ Imperfect Information Game ♦ Large-scale Multi-agent Problem ♦ Partial Observability ♦ Well-defined Example ♦ Computational Cost ♦ Expert-level Human Player ♦ Observable Multi-agent Problem ♦ Comparable Performance ♦ Rl Agent ♦ Sampling Technique
Abstract We present a model-based reinforcement learning (RL) scheme for large-scale multi-agent problems with partial observability, and apply it to the card game "Hearts", which is a well-defined example of an imperfect information game. To reduce the computational cost, we use a sampling technique based on Markov chain Monte Carlo (MCMC) in which the heavy integration required for the estimation and prediction can be approximated by a plausible number of samples. Computer simulation results show that our RL agent can perform learning of an appropriate strategy and exhibit a comparable performance to an expert-level human player in this partially observable multi-agent problem.
Educational Role Student ♦ Teacher
Age Range above 22 year
Educational Use Research
Education Level UG and PG ♦ Career/Technical Study