The generalization of AlphaGo Zero, called AlphaZero, achieves superhuman performance in all of Chess, Shogi, and Go. Starting from random play, and given no domain knowledge. New paper from DeepMind: https://arxiv.org/abs/1712.01815
-
Show this thread
-
To be fair, “no domain knowledge” is not accurate. The input features and network architecture are domain knowledge. Unclear how to generalize to domains that look very different, but at least the MCTS part is general purpose. Authors acknowledge this. https://twitter.com/dennybritz/status/938245388858925062 …pic.twitter.com/USpkLOdxec
3 replies 25 retweets 80 likesShow this thread
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.