AlphaGo Zero - AshokBhat/ml GitHub Wiki

About

• AI system developed by DeepMind
• Successor to AlphaGo, introduced in 2017
• Learned to play Go entirely through self-play
• Did not use human game data
• Uses deep learning and reinforcement learning
• Surpassed earlier versions of AlphaGo
• Its approach was later generalized in AlphaZero

See also

• [AlphaGo]] ](/AshokBhat/ml/wiki/[[AlphaZero)
• [Google DeepMind]] ](/AshokBhat/ml/wiki/[[Go)
• Reinforcement Learning