factual

AlphaGo Zero outperformed by learning only from self-play

AlphaGo Zero was trained exclusively through self-play using random moves and a reward signal rather than expert human games, and it outclassed its predecessor by a wide margin, discovering strategies unknown to human players.

factualpending

Speaker

David

Evidence Quote

“By unshackling itself from human wisdom, the model had discovered strategies unknown to mortal players.”

Source

#416: The Relentless Missionary Creating AGI: Demis Hassabis— Founders
Created: 11/12/2026, 7:25:16 PM

Your notes

Loading notes...