Learning to Play 7 Wonders Duel Without Human Supervision

TL;DR

This paper presents ZeusAI, an AI system that learns to play 7 Wonders Duel using reinforcement learning, combining Monte Carlo Tree Search and Transformer Neural Networks, achieving top human-level performance and exploring game strategies.

Contribution

Introduces ZeusAI, a novel AI framework for learning complex board games without human data, utilizing reinforcement learning with MCTS and Transformers.

Findings

ZeusAI reaches top human-level performance.

Develops both known and novel strategies.

Enables testing of game rule variants for balance.

Abstract

This paper introduces ZeusAI, an artificial intelligence system developed to play the board game 7 Wonders Duel. Inspired by the AlphaZero reinforcement learning algorithm, ZeusAI relies on a combination of Monte Carlo Tree Search and a Transformer Neural Network to learn the game without human supervision. ZeusAI competes at the level of top human players, develops both known and novel strategies, and allows us to test rule variants to improve the game's balance. This work demonstrates how AI can help in understanding and enhancing board games.