We present JueWu-SL, the first supervised-learning-based artificial intelligence (AI) program that achieves human-level performance in playing multiplayer online battle arena (MOBA) games. Unlike prior attempts, we integrate the macro-strategy and the micromanagement of MOBA-game-playing into neural networks in a supervised and end-to-end manner. Tested on Honor of Kings, the most popular MOBA at present, our AI performs competitively at the level of High King players in standard 5v5 games.
@article{arxiv.2011.12582,
title = {Supervised Learning Achieves Human-Level Performance in MOBA Games: A Case Study of Honor of Kings},
author = {Deheng Ye and Guibin Chen and Peilin Zhao and Fuhao Qiu and Bo Yuan and Wen Zhang and Sheng Chen and Mingfei Sun and Xiaoqian Li and Siqin Li and Jing Liang and Zhenjie Lian and Bei Shi and Liang Wang and Tengfei Shi and Qiang Fu and Wei Yang and Lanxiao Huang},
journal= {arXiv preprint arXiv:2011.12582},
year = {2020}
}
Comments
IEEE Transactions on Neural Networks and Learning Systems (TNNLS)