Reinforcement Learning for ConnectX

TL;DR

This paper explores the application of reinforcement learning algorithms to the complex, parameterized game ConnectX, which generalizes Connect 4 with variable board sizes and win conditions.

Contribution

The paper introduces reinforcement learning methods tailored for ConnectX, addressing its unique challenges and demonstrating their effectiveness in this generalized game setting.

Findings

Reinforcement learning algorithms can be adapted to play ConnectX effectively.

The methods outperform baseline strategies in various game configurations.

The approach demonstrates the potential for RL in complex, parameterized board games.

Abstract

ConnectX is a two-player game that generalizes the popular game Connect 4. The objective is to get X coins across a row, column, or diagonal of an M x N board. The first player to do so wins the game. The parameters (M, N, X) are allowed to change in each game, making ConnectX a novel and challenging problem. In this paper, we present our work on the implementation and modification of various reinforcement learning algorithms to play ConnectX.