Pareto-guided Pipeline for Distilling Featherweight AI Agents in Mobile MOBA Games

TL;DR

This paper presents a Pareto-guided pipeline for distilling large game AI models into lightweight, efficient agents suitable for mobile MOBA games, balancing performance and resource constraints.

Contribution

It introduces a systematic pipeline and a tailored search space for creating mobile-friendly AI agents that maintain high performance while significantly improving efficiency.

Findings

12.4x faster inference speed

15.6x better energy efficiency

40.32% win rate retention

Abstract

Recent advances in game AI have demonstrated the feasibility of training agents that surpass top-tier human professionals in complex environments such as Honor of Kings (HoK), a leading mobile multiplayer online battle arena (MOBA) game. However, deploying such powerful agents on mobile devices remains a major challenge. On one hand, the intricate multi-modal state representation and hierarchical action space of HoK demand large, sophisticated policy networks that are inherently difficult to compress into lightweight forms. On the other hand, production deployment requires high-frequency inference under strict energy and latency constraints on mobile platform. To the best of our knowledge, bridging large-scale game AI and practical on-device deployment has not been systematically studied. In this work, we propose a Pareto optimality guided pipeline and design a high-efficiency student architecture search space tailored for mobile execution, enabling systematic exploration of the trade-off between performance and efficiency. Experimental results demonstrate that the distilled model achieves remarkable efficiency, including an $12.4\times$ faster inference speed (under 0.5ms per frame) and a $15.6\times$ improvement in energy efficiency (under 0.5mAh per game), while retaining a 40.32% win rate against the original teacher model.