Evolutionary Transfer Learning for Dragonchess

TL;DR

This paper explores using evolutionary transfer learning to adapt chess heuristics for Dragonchess, demonstrating that evolutionary optimization enhances AI performance in complex, multi-layered game environments.

Contribution

It introduces Dragonchess as a new AI research testbed and applies evolutionary transfer learning to adapt existing heuristics, showing significant performance improvements.

Findings

Evolutionary optimization improved AI performance in Dragonchess.

Direct heuristic transfer was ineffective due to structural differences.

Dragonchess serves as a novel complex domain for AI research.

Abstract

Dragonchess, a three-dimensional chess variant introduced by Gary Gygax, presents unique strategic and computational challenges that make it an ideal environment for studying the transfer of artificial intelligence (AI) heuristics across domains. In this work, we introduce Dragonchess as a novel testbed for AI research and provide an open-source, Python-based game engine for community use. Our research investigates evolutionary transfer learning by adapting heuristic evaluation functions directly from Stockfish, a leading chess engine, and subsequently optimizing them using Covariance Matrix Adaptation Evolution Strategy (CMA-ES). Initial trials showed that direct heuristic transfers were inadequate due to Dragonchess's distinct multi-layer structure and movement rules. However, evolutionary optimization significantly improved AI agent performance, resulting in superior gameplay demonstrated through empirical evaluation in a 50-round Swiss-style tournament. This research establishes the effectiveness of evolutionary methods in adapting heuristic knowledge to structurally complex, previously unexplored game domains.