AlphaGo Zero

In 2016, he said, "AlphaGo beat Lee Sedol in a milestone for AI, but key to that was AI's ability to, quote, 'ponder' for one minute in, uh, before each move.

How much did that improve it? For AlphaGo Zero, it's the equivalent of a scaling pre-training by 100,000x." Then he said, "All those prior methods are specific to that game.

But if we can discover a general version, the benefits could be huge.

Yes, inference may be 1,000 times slower and more costly," meaning let- letting the model think before it responds.

Mm-hmm.

But the later versions, like AlphaGo Zero, they didn't learn from humans at all.

And that represents the ultimate paradigm shift in the field.

AlphaGo Zero started with completely blank slate.

It only knew the rules of how a stone could be placed.

It generated its own training data by playing millions of games against itself.