# Similarity Measures based on Local Game Trees

## TL;DR

This paper introduces similarity measures based on local game tree structures to improve detection of forcing continuations in two-player perfect information games, enhancing game-playing agents.

## Contribution

It proposes new similarity measures derived from local game trees and evaluates their effectiveness in matching trap states in chess.

## Key findings

- Measures show promising accuracy in identifying trap states.
- Improves detection of forcing continuations in game-playing agents.
- Provides a new approach to analyzing game positions using local tree structures.

## Abstract

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.