+++ Update 10/2020: FutureWorkGBG: Project themes in GBG for students (PDF) +++
Games are an interesting field in computer science concerning the question whether a computer can learn the game strategies just from self-play, without explicitly programming the tactics or performing exhaustive search. This is a branch of artificial intelligence (AI).
GBG takes the abstraction one level higher (than in the previous Connect-4 project) in that it provides a software framework with standardized interfaces for arbitrary games and arbitrary AI agents. GBG helps students and researchers to take a quicker start-off into the area of game learning. Read more about GBG in the publications below.
Currently, the games implemented in GBG include 2048, Hex, Othello, Sim, Nim, Tic-Tac-Toe; more games are planned for the future. Current agents in GBG include MCTS, Max-N, Expectimax-N, TD-n-tuple and others.
The long-term goal of our research group is it to transfer these learning strategies to many other games (dots-and-boxes, go, Poker, checkers, Abalone, ...). The project is related to the research field known as General Game Playing (GGP). The aim of GGP and GBG is it to develop agents which are able to learn a great variety of games.
Read more about Games & Learning ...
The article General Board Game Playing for Education and Research in Generic AI Game Learning (2019) provides a scientific overview on GBG, the technical report The GBG Class Interface Tutorial V2.2: General Board Game Playing and Learning (Oct 2020) gives an introduction to GBG more from the programmer's perspective.