AlphaGo Zero And AlphaZero, RomiChess done better

[Ovyron]

It probably can be done without tying up too many resources, since the engines that have learning features are quite few - Critter, Baron, Phalanx, RomiChess, of course, and maybe a few others.

Critter's learning is flawed because it's blind unless it reaches more depth, which eventually becomes prohibitive. It also has problems learning about refutations at low depth, which are very important (Say, Critter searches up to depth 24 some line and makes a move, the user knows this move loses and plays the refutation, Critter sees this new line is best since depth 10, but to see it from the root would require reaching depth 25, which isn't practical so Critter may lose this position over and over.)

Stockfish (PA_GTB) learning is flawed because it's very easy to confuse it in some endgame, make it show 0.00, backtrack this to the root, in a lost position. If this happens by accident Stockfish will aim to this position thinking it's a draw and lose it every time.

Other engines with Learning:

Yace - No comment on this one. The engine was so weak that I couldn't check the effect of learning. Though it's stronger than RomiChess

Rybka 3 - Has actually a very good learning. The problem is the user needs to manually propagate it, so no way you're going to be playing 1000000 games with it, as you'd need to sit after every game and make sure Rybka learns.

Shredder 11 - My favorite learning algorythm thus far. Doesn't have the flaw of other learning algorythms and one can jump around more easily. I think it's limited to a small file, though. And it was removed in Shredder 12 or 13 (never had 12, but 13 doesn't have the feature.)

Edit - Oh, and don't forget Houdini's learning, which was also removed on later versions. In my experience its learning was useless.

[Ras]

And it was removed in Shredder 12 or 13 (never had 12, but 13 doesn't have the feature.)

Shredder 13: Click Menu -> Extras -> Options, then in the lower left corner "Book Options". In the new dialog in the upper right corner, I can see "Book Learning" and "Result Learning". You can enable these if you have "Use Opening Book" checked.

Or isn't that what you are referring to?

[Ovyron]

Shredder 13: Click Menu -> Extras -> Options, then in the lower left corner "Book Options". In the new dialog in the upper right corner, I can see "Book Learning" and "Result Learning". You can enable these if you have "Use Opening Book" checked.

Or isn't that what you are referring to?

What you're describing is a GUI feature for books, which all engines can use, that modifies the weights on the book's moves so that losing moves will be played less often and so on, similar to what the Fritz GUI does with CTG books.

What I'm describing is "Position Learning", a feature that Shredder 11 has (under Extras > Engine > Engine Options), that Shredder 13 does not, that allows it to learn scores of positions to use them on future games that reach those or similar positions. A great feature for analysis of games, as the engine can see refutations of lines and suggest different (and better) plans for the same position as you go forward and backward in the analysis, even after unloading and loading the engine.

It's a shame it was removed, though, Shredder 11 was terrible for analysis even with it, and Shredder 13 is really good even without it, though it seems like a new engine altogether, it seems like a good tradeoff.