3
$\begingroup$

Google says that their new AI program MuZero learnt the rules of chess and some other board games without being told so. How is this even possible?

https://towardsdatascience.com/deepmind-unveils-muzero-a-new-agent-that-mastered-chess-shogi-atari-and-go-without-knowing-the-d755dc80ff08

$\endgroup$
  • 3
    $\begingroup$ Check Appendix A of the original paper, section titled "Actions Available". The rules are actually used at the root of the search tree. But after that, the hidden state is evaluated independent of the rules; the model learns to make legal moves. arxiv.org/pdf/1911.08265.pdf $\endgroup$ – serali Nov 30 '19 at 12:55
  • 1
    $\begingroup$ Why do you think this is impossible? You could ask the exact same question about any feat of human learning. If you are interested in the specific algorithms that are used, you should clarify in your question. $\endgroup$ – Mathias Müller Jan 28 at 17:35

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.