![Mastering chess and shogi by self-play with a general reinforcement learning algorithm | the morning paper Mastering chess and shogi by self-play with a general reinforcement learning algorithm | the morning paper](https://blog.acolyer.org/wp-content/uploads/2018/01/alphazero-table-1.jpeg?w=520)
Mastering chess and shogi by self-play with a general reinforcement learning algorithm | the morning paper
![How to build your own AlphaZero AI using Python and Keras | by David Foster | Applied Data Science | Medium How to build your own AlphaZero AI using Python and Keras | by David Foster | Applied Data Science | Medium](https://miro.medium.com/max/1400/1*PanQe3XaV_tpFLddTzpz3g.png)
How to build your own AlphaZero AI using Python and Keras | by David Foster | Applied Data Science | Medium
![To those saying it was unfair that Stockfish vs AlphaZero was played with very fast time controls: AlphaZero actually scales better than stockfish with time, suggesting that matches with longer time controls To those saying it was unfair that Stockfish vs AlphaZero was played with very fast time controls: AlphaZero actually scales better than stockfish with time, suggesting that matches with longer time controls](https://external-preview.redd.it/7e-C7VrGOOWbIftFCmvu-gJ86ztGUBWyHBkdpwEaMWc.jpg?auto=webp&s=825f6809aeb516254e274c3ec4cd341458e74f0b)
To those saying it was unfair that Stockfish vs AlphaZero was played with very fast time controls: AlphaZero actually scales better than stockfish with time, suggesting that matches with longer time controls
![Training AlphaZero for 700,000 steps. Elo ratings were computed from... | Download Scientific Diagram Training AlphaZero for 700,000 steps. Elo ratings were computed from... | Download Scientific Diagram](https://www.researchgate.net/publication/321571298/figure/fig1/AS:568407281147904@1512530264876/Training-AlphaZero-for-700-000-steps-Elo-ratings-were-computed-from-evaluation-games.png)