HackerPapers

arXiv:2111.09259cs.AI

Acquisition of Chess Knowledge in AlphaZero

Thomas McGrath, Andrei Kapishnikov, Nenad Tomašev, Adam Pearce, Demis Hassabis, Been Kim, Ulrich Paquet, Vladimir Kramnik

Abstract

What is learned by sophisticated neural network agents such as AlphaZero? This question is of both scientific and practical interest. If the representations of strong neural networks bear no resemblance to human concepts, our ability to understand faithful explanations of their decisions will be restricted, ultimately limiting what we can achieve with neural network interpretability. In this work we provide evidence that human knowledge is acquired by the AlphaZero neural network as it trains on the game of chess. By probing for a broad range of human chess concepts we show when and where these concepts are represented in the AlphaZero network. We also provide a behavioural analysis focusing on opening play, including qualitative analysis from chess Grandmaster Vladimir Kramnik. Finally, we carry out a preliminary investigation looking at the low-level details of AlphaZero's representations, and make the resulting behavioural and representational analyses available online.

arXiv comments
69 pages, 44 figures
Journal reference
DOI
doi:10.1073/pnas.2206625119
Subjects
cs.AIArtificial Intelligence stat.MLMachine Learning (Statistics)

On Hacker News

Linked by 5 identifiable commenters in 5 comments across 4 threads, from Nov 2021 to Feb 2026. Found in 5 HN submissions from Nov 2021 to Nov 2021. Open submission discussions.

Comments linking this paper, per year.

Discussed in comments4 threads

  1. Experts Have World Models. LLMs Have Word Models 1 comment

    • > AlphaGo or AlphaZero didn’t need to model human cognition. It needed to see the current state and calculate the optimal path better than any human could. I don't think this is right: To calculate the optimal path, you do need to model human cognition. At least, in the sense…

      azakaiView on HNParent context

  2. Acquisition of chess knowledge in AlphaZero 1 comment

    • Anyone know how this differs from a similar-seeming paper that was published a year ago? https://en.chessbase.com/post/acquisition-of-chess-knowledge... https://arxiv.org/pdf/2111.09259.pdf

      dsjoergView on HNParent context

  3. Can deep learning help mathematicians build intuition? 1 comment

    • It looks like the basic meta-algorithm is: 1) Notice a pattern in the mathematics 2) Device a neural network that leverages structural properties of the space you wish to investigate 3) Encode relationship in a manner allowing use of supervised learning and see if the net can…

      VetchView on HNParent context

  4. Acquisition of chess knowledge in AlphaZero 2 comments

    • The paper itself is here: https://arxiv.org/abs/2111.09259 with the key conclusion that "Examining the evolution of human concepts using probing showed that many human concepts can be accurately regressed from the AZ network after training, even though AlphaZero has never seen a…

      saroshView on HNParent context

    • Why link to Chessbase instead of to the arxiv paper directly? https://arxiv.org/abs/2111.09259 Chessbase is notoriously sketchy and there is a lawsuit against them for blatantly stealing GPL code. https://stockfishchess.org/blog/2021/our-lawsuit-against-che...

      porphyraView on HNParent context

Submitted to Hacker News

Connected through shared HN threads, not inferred similarity or a recommendation.