DeepMind's AlphaZero has managed to learn and play three complex games, and won(eyerys.com)

https://www.eyerys.com/articles/timeline/deepminds-alphazero-ai-masters-chess-shogi-snd-go-through-self-play