NeuroPAL: Punctuated Anytime Learning with Neuroevolution for Macromanagement in Starcraft: Brood War

TL;DR

NeuroPAL combines neuroevolution and punctuated anytime learning to efficiently train AI agents for macromanagement in StarCraft: Brood War, achieving faster learning and emergent strategic behaviors.

Contribution

This work introduces NeuroPAL, a novel neuroevolutionary framework that integrates PAL with NEAT to improve training efficiency in complex RTS environments.

Findings

PAL accelerates training, halving the time needed for effective strategies.

Evolved agents demonstrate human-like strategic behaviors.

NeuroPAL outperforms standard NEAT in sample efficiency.

Abstract

StarCraft: Brood War remains a challenging benchmark for artificial intelligence research, particularly in the domain of macromanagement, where long-term strategic planning is required. Traditional approaches to StarCraft AI rely on rule-based systems or supervised deep learning, both of which face limitations in adaptability and computational efficiency. In this work, we introduce NeuroPAL, a neuroevolutionary framework that integrates Neuroevolution of Augmenting Topologies (NEAT) with Punctuated Anytime Learning (PAL) to improve the efficiency of evolutionary training. By alternating between frequent, low-fidelity training and periodic, high-fidelity evaluations, PAL enhances the sample efficiency of NEAT, enabling agents to discover effective strategies in fewer training iterations. We evaluate NeuroPAL in a fixed-map, single-race scenario in StarCraft: Brood War and compare its performance to standard NEAT-based training. Our results show that PAL significantly accelerates the learning process, allowing the agent to reach competitive levels of play in approximately half the training time required by NEAT alone. Additionally, the evolved agents exhibit emergent behaviors such as proxy barracks placement and defensive building optimization, strategies commonly used by expert human players. These findings suggest that structured evaluation mechanisms like PAL can enhance the scalability and effectiveness of neuroevolution in complex real-time strategy environments.