Similarity Measures based on Local Game Trees
This work addresses the challenge of improving game-playing agents' ability to identify forcing continuations, but it appears incremental as it focuses on specific similarity measures for a narrow domain.
The paper tackled the problem of detecting strategic similarity in game positions for two-player perfect-information games by analyzing local game tree structures, resulting in promising accuracy in matching trap states in chess benchmarks.
We study strategic similarity of game positions in two-player extensive games of perfect information, by looking at the structure of their local game trees, with the aim of improving the performance of game playing agents in detecting forcing continuations. We present a range of measures over the induced game trees and compare them against benchmark problems in chess, observing a promising level of accuracy in matching up trap states.