Acquisition of Chess Knowledge in AlphaZero

Re:Acquisition of Chess Knowledge in AlphaZero by Waterluvian on Sunday November 20, 2022 @02:05PM (#33684071)

Does AI still struggle with “I can’t tell you how I derived this answer”? Is that improving much?

[ Parent | on HN ]

Re:Acquisition of Chess Knowledge in AlphaZero by Tenoke on Sunday November 20, 2022 @02:08PM (#33684103)

Yes exploitability is improving but this hasn't been that much of a problem with top chess engines at any point.

[ Parent | on HN ]

Re:Acquisition of Chess Knowledge in AlphaZero by firsttimebigboy on Sunday November 20, 2022 @06:09PM (#33686529)

AI doesn't even attempt that. All our neural networks do is spit out the final result of whatever it's doing. If we want the "reasoning" for it we have to build different methods of doing that. It's an active field of research but much slower going than improving our AI's directly.

[ Parent | on HN ]