让强模型在手机上跑得又快又省电，还能赢40%比赛。
Pareto-guided Pipeline for Distilling Featherweight AI Agents in Mobile MOBA Games
- 用帕累托优化框架找性能与效率的最佳平衡点。
- 推理速度提升12.4倍（每帧<0.5ms），能耗降15.6倍（每局<0.5mAh）。
- 专为手机设计轻量架构，适合部署高阶游戏AI到移动端。
近年来，游戏AI在《荣耀》（Honor of Kings, HoK）等复杂移动端MOBA游戏中已实现超越顶级人类选手的水平。然而，将此类强大智能体部署至移动设备仍面临巨大挑战：一方面，HoK的多模态状态表示与分层动作空间要求庞大复杂的策略网络，难以压缩为轻量形式；另一方面，实际部署需满足移动平台严格的能耗与延迟约束，实现高频推理。据我们所知，尚未有系统性研究解决大规模游戏AI与实际设备部署之间的鸿沟。本文提出一种基于帕累托最优的蒸馏流程，并设计适配移动端执行的高效学生网络搜索空间，系统探索性能与效率的权衡。实验表明，蒸馏后模型在保持40.32%对原教师模型胜率的同时，推理速度提升12.4倍（每帧<0.5ms），能耗降低15.6倍（每局<0.5mAh）。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。