As artificial intelligence becomes increasingly intelligent—in some cases, achieving superhuman performance—there is growing potential for humans to learn from and collaborate with algorithms. However, the ways in which AI systems approach problems are often different from the ways people do, and thus may be uninterpretable and hard to learn from. A crucial step in bridging this gap between human and artificial intelligence is modeling the granular actions that constitute human behavior, rather than simply matching aggregate human performance.

We pursue this goal in a model system with a long history in artificial intelligence: chess. The aggregate performance of a chess player unfolds as they make decisions over the course of a game. The hundreds of millions of games played online by players at every skill level form a rich source of data in which these decisions, and their exact context, are recorded in minute detail. Applying existing chess engines to this data, including an open-source implementation of AlphaZero, we find that they do not predict human moves well.

We develop and introduce Maia, a customized version of AlphaZero trained on human chess games, that predicts human moves at a much higher accuracy than existing engines, and can achieve maximum accuracy when predicting decisions made by players at a specific skill level in a tuneable way. For a dual task of predicting whether a human will make a large mistake on the next move, we develop a deep neural network that significantly outperforms competitive baselines. Taken together, our results suggest that there is substantial promise in designing artificial intelligence systems with human collaboration in mind by first accurately modeling granular human decision-making.

Blog posts

Introducing Maia: a Human-Like Chess Engine
We develop and introduce Maia, a machine learning model that captures human style in chess.

Publications

Learning Models of Individual Behavior in Chess
Reid McIlroy-Young, Russell Wang, Siddhartha Sen, Jon Kleinberg, and Ashton Anderson. KDD 2022.
BibTeX
@inproceedings{mcilroy2022learning,
  title={Learning models of individual behavior in chess},
  author={McIlroy-Young, Reid and Wang, Russell and Sen, Siddhartha and Kleinberg, Jon and Anderson, Ashton},
  booktitle={Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining},
  pages={1253--1263},
  year={2022}
}
Detecting Individual Decision-Making Style: Exploring Behavioral Stylometry in Chess
Reid McIlroy-Young, Russell Wang, Siddhartha Sen, Jon Kleinberg, Ashton Anderson. NeurIPS 2021.
BibTeX
@article{mcilroy2021detecting,
  title={Detecting individual decision-making style: Exploring behavioral stylometry in chess},
  author={McIlroy-Young, Reid and Wang, Yu and Sen, Siddhartha and Kleinberg, Jon and Anderson, Ashton},
  journal={Advances in Neural Information Processing Systems},
  volume={34},
  pages={24482--24497},
  year={2021}
}
Learning Personalized Models of Human Behavior in Chess
Reid McIlroy-Young, Russell Wang, Siddhartha Sen, Jon Kleinberg, Ashton Anderson. Working paper.
BibTeX
@article{mcilroy2020learning,
  title={Learning Personalized Models of Human Behavior in Chess},
  author={McIlroy-Young, Reid and Wang, Russell and Sen, Siddhartha and Kleinberg, Jon and Anderson, Ashton},
  journal={arXiv preprint arXiv:2008.10086},
  year={2020}
}
Aligning Superhuman AI with Human Behavior: Chess as a Model System
Reid McIlroy-Young, Siddhartha Sen, Jon Kleinberg, Ashton Anderson. KDD 2020.
BibTeX
@article{mcilroyyoung2020maia,
  title={Aligning Superhuman AI with Human Behavior: Chess as a Model System},
  author={McIlroy-Young, Reid and  Sen, Siddhartha and Kleinberg, Jon and Anderson, Ashton},
  year={2020},
  eprint={2006.01855},
  archivePrefix={arXiv},
  primaryClass={cs.AI}
}