Mastering Complex Control in MOBA Games with Deep Reinforcement Learning

  • Deheng Ye Tencent
  • Zhao Liu Tencent
  • Mingfei Sun Tencent
  • Bei Shi Tencent
  • Peilin Zhao Tencent
  • Hao Wu Tencent
  • Hongsheng Yu Tencent
  • Shaojie Yang Tencent
  • Xipeng Wu Tencent
  • Qingwei Guo Tencent
  • Qiaobo Chen Tencent
  • Yinyuting Yin Tencent
  • Hao Zhang Tencent
  • Tengfei Shi Tencent
  • Liang Wang Tencent
  • Qiang Fu Tencent
  • Wei Yang Tencent
  • Lanxiao Huang Tencent

Abstract

We study the reinforcement learning problem of complex action control in the Multi-player Online Battle Arena (MOBA) 1v1 games. This problem involves far more complicated state and action spaces than those of traditional 1v1 games, such as Go and Atari series, which makes it very difficult to search any policies with human-level performance. In this paper, we present a deep reinforcement learning framework to tackle this problem from the perspectives of both system and algorithm. Our system is of low coupling and high scalability, which enables efficient explorations at large scale. Our algorithm includes several novel strategies, including control dependency decoupling, action mask, target attention, and dual-clip PPO, with which our proposed actor-critic network can be effectively trained in our system. Tested on the MOBA game Honor of Kings, the trained AI agents can defeat top professional human players in full 1v1 games.

Published
2020-04-03
Section
AAAI Technical Track: Machine Learning