Elo World, a framework forbenchmarking weak chess engines

Discussion of chess software programming and technical issues.

Moderators: bob, hgm, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Post Reply
Rémi Coulom
Posts: 434
Joined: Mon Apr 24, 2006 6:06 pm
Contact:

Elo World, a framework forbenchmarking weak chess engines

Post by Rémi Coulom » Tue Apr 02, 2019 9:53 am

interesting paper:
http://tom7.org/chess/weak.pdf

the Association for Computational Heresy has more of this kind:
http://sigbovik.org/

Ferdy
Posts: 4111
Joined: Sun Aug 10, 2008 1:15 pm
Location: Philippines

Re: Elo World, a framework forbenchmarking weak chess engines

Post by Ferdy » Tue Apr 02, 2019 10:47 am

Rémi Coulom wrote:
Tue Apr 02, 2019 9:53 am
interesting paper:
http://tom7.org/chess/weak.pdf

the Association for Computational Heresy has more of this kind:
http://sigbovik.org/
I like the Pacifist and Generous methods, both actually tops the rating list 2nd and 4th respectively :).
Will use these in one of the versions of CDrill.

User avatar
Laskos
Posts: 9467
Joined: Wed Jul 26, 2006 8:21 pm
Full name: Kai Laskos

Re: Elo World, a framework forbenchmarking weak chess engines

Post by Laskos » Tue Apr 02, 2019 12:40 pm

Rémi Coulom wrote:
Tue Apr 02, 2019 9:53 am
interesting paper:
http://tom7.org/chess/weak.pdf

the Association for Computational Heresy has more of this kind:
http://sigbovik.org/
Sorry for being dumb, but this p(champion) probability of the first paper derived from Markov discrete-time chain matrix is that at current (finishing) time of his tourney? I mean, if he had a 10 times longer tourney, the values of p in the table would diverge somewhat from smallest to largest values? So, bearing close relation to LOS matrix?

Daniel Shawul
Posts: 3758
Joined: Tue Mar 14, 2006 10:34 am
Location: Ethiopia
Contact:

Re: Elo World, a framework forbenchmarking weak chess engines

Post by Daniel Shawul » Tue Apr 02, 2019 3:28 pm

Reading this paper, my feeling goes like: He is joking, no he is serious, haha april fool joking, that is actually a good point, this is crap ...
I can't decide.

User avatar
Laskos
Posts: 9467
Joined: Wed Jul 26, 2006 8:21 pm
Full name: Kai Laskos

Re: Elo World, a framework forbenchmarking weak chess engines

Post by Laskos » Tue Apr 02, 2019 3:50 pm

Daniel Shawul wrote:
Tue Apr 02, 2019 3:28 pm
Reading this paper, my feeling goes like: He is joking, no he is serious, haha april fool joking, that is actually a good point, this is crap ...
I can't decide.
Yeah, I feel the same, but AFTER reading your post. I was thinking that I just cannot collect myself to read this thing :lol:.

Albert Silver
Posts: 2861
Joined: Wed Mar 08, 2006 8:57 pm
Location: Rio de Janeiro, Brazil

Re: Elo World, a framework forbenchmarking weak chess engines

Post by Albert Silver » Tue Apr 02, 2019 4:20 pm

Laskos wrote:
Tue Apr 02, 2019 3:50 pm
Daniel Shawul wrote:
Tue Apr 02, 2019 3:28 pm
Reading this paper, my feeling goes like: He is joking, no he is serious, haha april fool joking, that is actually a good point, this is crap ...
I can't decide.
Yeah, I feel the same, but AFTER reading your post. I was thinking that I just cannot collect myself to read this thing :lol:.
It is a joke you guys. If it helps, read the last page:

"The author would also like to thank the anonymous referees of the Special Interest Group on Bafflingly Overdone Ventures In Chess journal for their careful reviews."
"Tactics are the bricks and sticks that make up a game, but positional play is the architectural blueprint."


konsolas
Posts: 182
Joined: Sun Jun 12, 2016 3:44 pm
Location: London
Full name: Vincent
Contact:

Re: Elo World, a framework forbenchmarking weak chess engines

Post by konsolas » Tue Apr 02, 2019 10:10 pm

I'm pleasantly surprised that Topple turned up in that paper. It's pretty funny too :)

Sesse
Posts: 180
Joined: Mon Apr 30, 2018 9:51 pm
Contact:

Re: Elo World, a framework forbenchmarking weak chess engines

Post by Sesse » Fri Apr 05, 2019 3:57 pm

Albert Silver wrote:
Tue Apr 02, 2019 4:20 pm
It is a joke you guys. If it helps, read the last page:

"The author would also like to thank the anonymous referees of the Special Interest Group on Bafflingly Overdone Ventures In Chess journal for their careful reviews."
It's a SIGBOVIK paper; SIGBOVIK is a joke conference. But Tom7 frequently submits interesting-but-strange research to SIGBOVIK, so it's not a fake paper. He actually went through all of this to get a humourous result (and paper). Look up his previous stuff; he's done generic self-learning NES algorithms, autoconverting old games to 3D, a C89 compiler that compiles to readable ASCII (where the paper is in itself compilable), and much more.

He actually submitted multiple SIGBOVIK papers about chess this year. One of them covers known ground (which pieces are the most likely to survive, and what pieces on the chess board are the bloodiest), but I found the one about playing piece-blind chess interesting:

http://radar.spacebar.org/f/a/weblog/comment/1/1167

It turns out you can get to >500 Elo just by knowing where there are pieces, making a guess at which pieces those are using a neural network, and then going from there.

AlvaroBegue
Posts: 921
Joined: Tue Mar 09, 2010 2:46 pm
Location: New York
Full name: Álvaro Begué (RuyDos)

Re: Elo World, a framework forbenchmarking weak chess engines

Post by AlvaroBegue » Sun Jul 21, 2019 11:22 pm

'Regarding this thread, I just found this fascinating video on YouTube: https://www.youtube.com/watch?v=DpXy041BIlA

Enjoy!
Álvaro.

Post Reply