DeepMind introduced AlphaZero in 2017. It is a single system that taught itself how to master chess, shogi, and Go, beating state-of-the-art programs in each case. AlphaZero has developed a ground-breaking, highly dynamic, and unconventional style of play.

A report titled: “A general reinforcement learning algorightm that masters chess, shogi an Go through self-play” was published by Science Magazine. Part of the report said: “The ability of AlphaZero to adapt to various game rules is a notable step toward achieving a general game-playing system.”

AlphaZero replaces the handcrafted knowledge and domain-specific augmentations used in traditional game-playing programs with deep neural networks, a general-purpose reinforcement learning algorithm, and a general-purpose tree search algorithm.

In chess, AlphaZero first outperformed Stockfish after just 4 hours. AlphaZero defeated the 2016 TCEC (Season 9) world champion Stockfish, winning 155 games and losing just six games out of 1,000.

In shogi, AlphaZero first outperformed Elmo after 2 hours. AlphaZero defeated the 2017 CSA world champion version of Elmo, winning 91.2% of games.

In Go, AlphaZero defeated AlphaGo Zero, winning 61% of games.

I wonder if, in the future, eSports will include competitions between AlphaZero and various other AI algorithms. It seems to me that people who love to play chess are very interested in AlphaZero and what it can do. I can see potential for chess players to learn some of AlphaZero’s strategies in an effort to improve their game.