Search found 147 matches

by clumma
Tue Dec 19, 2017 12:18 am
Forum: Computer Chess Club: General Topics
Topic: Cerebellum analysis of the AlphaZero - Stockfish Games
Replies: 21
Views: 6074

Re: Cerebellum analysis of the AlphaZero - Stockfish Games

Thank you, Zipproth. I was wondering what Cerebellum would say about these games.

I would also like to say, I am ready to buy Cerebellum, even if all features are not available yet. And please PM me if you need a beta tester!

-Carl
by clumma
Sat Dec 09, 2017 5:58 pm
Forum: Computer Chess Club: General Topics
Topic: Alpha Zero vs Stockfish 8 tournament conditions.
Replies: 82
Views: 17088

Re: Alpha Zero vs Stockfish 8 tournament conditions.

You are completely mistaken if you think that Google threw half a computing centre at the match. In the self-training, yes, that was a different story, but so was Stockfish's development (e.g. fishtest). In particular, they played 44 million games in 9 hours. Fishtest does that in about 3 weeks. -C...
by clumma
Fri Dec 08, 2017 9:36 pm
Forum: Computer Chess Club: General Topics
Topic: AlphaZero beats AlphaGo Zero, Stockfish, and Elmo
Replies: 198
Views: 46049

Re: AlphaZero beats AlphaGo Zero, Stockfish, and Elmo

lkaufman wrote:I agree that the test was far from fair; another issue that hurt SF was that it was apparently run on 64 threads on a 32 core machine,
Source? I don't see this info in the paper.

-Carl
by clumma
Fri Dec 08, 2017 8:57 pm
Forum: Computer Chess Club: General Topics
Topic: Historic Milestone: AlphaZero
Replies: 193
Views: 70024

Re: Historic Milestone: AlphaZero

My crude estimate goes... * Shredder 12 won WCSC 2015 on 4 cores with 45 min and a 15s increment. Regan estimates its FIDE Elo in this tournament at 3075. * On CCRL 40/40 list, Shredder 12 4CPU Elo is 3025. So this list is maybe 50 Elo below FIDE scale. But this is within Regan's margin of error an...
by clumma
Fri Dec 08, 2017 4:02 am
Forum: Computer Chess Club: General Topics
Topic: Historic Milestone: AlphaZero
Replies: 193
Views: 70024

Re: Historic Milestone: AlphaZero

AlphaZero's total score against SF 8 seems to be 797 out of 1300 (318/958/24), which is +80 Elo.
by clumma
Fri Dec 08, 2017 3:44 am
Forum: Computer Chess Club: General Topics
Topic: Alpha Zero vs Stockfish 8 tournament conditions.
Replies: 82
Views: 17088

Re: Alpha Zero vs Stockfish 8 tournament conditions.

The only thing I question is that AlphaZero had an opening book based on learning over millions of games, whereas it looks as though Stockfish may not have used an opening book at all. Stockfish was fine out of the opening (in the 10 games at 1 min/move). AlphaZero also played 1200 games against SF...
by clumma
Fri Dec 08, 2017 2:03 am
Forum: Computer Chess Club: General Topics
Topic: Historic Milestone: AlphaZero
Replies: 193
Views: 70024

Re: Historic Milestone: AlphaZero

Kai, can you estimate the FIDE-equivalent Elo for Stockfish 8 with 64 cores and 1 min per move? My crude estimate goes... * Shredder 12 won WCSC 2015 on 4 cores with 45 min and a 15s increment. Regan estimates its FIDE Elo in this tournament at 3075. * On CCRL 40/40 list, Shredder 12 4CPU Elo is 302...
by clumma
Fri Dec 08, 2017 12:50 am
Forum: Computer Chess Club: General Topics
Topic: AlphaZero beats AlphaGo Zero, Stockfish, and Elmo
Replies: 198
Views: 46049

Re: AlphaZero beats AlphaGo Zero, Stockfish, and Elmo

Y-scale in Fig.2 means Elo without added reference point (that is what that clumsy term google used actually means). It doesn't mean Elo difference between 2 programs, because if it was Elo difference you'd need a single curve, not 2 of them (it's pretty obvious if you just used your head a bit). T...
by clumma
Thu Dec 07, 2017 12:39 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: @Mathew Lai
Replies: 1
Views: 1047

Re: @Mathew Lai

Seconded!
by clumma
Wed Dec 06, 2017 6:17 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Google's AlphaGo team has been working on chess
Replies: 203
Views: 60189

Re: Google's AlphaGo team has been working on chess

Time is misleading in DeepMind's papers, as they use thousands of "computers" (not even commercially available). Money would be a better measure. Neural nets scale well on cheap (per FLOP) SIMD hardware, so you'll definitely pay more for the traditional engine at the same level of performance. -Carl