Seminar: Monte Carlo Tree Search for the Imperfect Information Card Game Cribbage

Richard Kelly
Honours Project
Supervisor: Dr. David Churchill

Monte Carlo Tree Search for the Imperfect Information Card Game Cribbage

Department of Computer Science
Tuesday, April 4, 2017, 12:00 p.m., Room EN-2022


Abstract

Monte Carlo Tree Search (MCTS), an AI technique that uses random sampling of game playouts, has been used successfully in perfect information deterministic games such as Go since 2006. It is a useful game AI technique because it doesn't rely on evaluations of non-terminal game states or other domain knowledge.

Several techniques for handling imperfect information and stochastic game elements in MCTS have been used. In this research, we implemented a Cribbage card game for two players and investigated several methods for handling imperfect information for an MCTS AI player.

 

Contact

Department of Computer Science

230 Elizabeth Ave

St. John's, NL A1B 3X9 CANADA

Tel: (709) 864-2530

Fax: (709) 864-2552

becomestudent@mun.ca