Mastering Complex Control in MOBA Games with Deep Reinforcement Learning

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, our AI agent, called Tencent Solo, can defeat top professional human players in full 1v1 games.

arXiv commentAAAI 2020

References in corpus (11)

• High-Dimensional Continuous Control Using Generalized Advantage Estimation
• Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm
• Human-level performance in first-person multiplayer games with population-based deep reinforcement learning
• IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures
• Horovod: fast and easy distributed deep learning in TensorFlow
• Emergent Complexity via Multi-Agent Competition
• An Empirical Model of Large-Batch Training
• Playing FPS Games with Deep Reinforcement Learning
• Learning to Communicate to Solve Riddles with Deep Distributed Recurrent Q-Networks
• MOBA: a New Arena for Game AI
• Marginal Policy Gradients: A Unified Family of Estimators for Bounded Action Spaces with Applications

Cited by in corpus (5)

• A Closer Look at Invalid Action Masking in Policy Gradient Algorithms
• Non-local Policy Optimization via Diversity-regularized Collaborative Exploration
• Policy Gradient from Demonstration and Curiosity
• Compare and Select: Video Summarization with Multi-Agent Reinforcement Learning
• Battlesnake Challenge: A Multi-agent Reinforcement Learning Playground with Human-in-the-loop