Monte Carlo Tree Search for a single target search game on a 2-D lattice

TL;DR

This paper investigates the effectiveness of Monte Carlo Tree Search in a 2-D lattice search game, comparing it to Levy Flight Search and analyzing its convergence properties.

Contribution

It introduces a novel application of MCTS to a 2-D search game and provides theoretical convergence proofs under idealized conditions.

Findings

MCTS performs competitively with Levy Flight Search.

Theoretical convergence of MCTS is established without computational constraints.

Analysis of different target distributions influences search efficiency.

Abstract

Monte Carlo Tree Search (MCTS) is a branch of stochastic modeling that utilizes decision trees for optimization, mostly applied to artificial intelligence (AI) game players. This project imagines a game in which an AI player searches for a stationary target within a 2-D lattice. We analyze its behavior with different target distributions and compare its efficiency to the Levy Flight Search, a model for animal foraging behavior. In addition to simulated data analysis we prove two theorems about the convergence of MCTS when computation constraints neglected.