Random playout vs evaluation

Discussion of chess software programming and technical issues.

Moderators: Harvey Williamson, bob, hgm

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Post Reply
Robert Pope
Posts: 523
Joined: Sat Mar 25, 2006 7:27 pm

Random playout vs evaluation

Post by Robert Pope » Fri May 15, 2015 6:47 pm

SJE's posts got me thinking.

How strong a correlation there would be between a monte carlo playout of a position vs. a traditional search? It seems that monte carlo would perform poorly when there is a narrow forced win. But perhaps the reverse is true, and random playout could identify positions where an engine's evaluation has a blind spot.

Has anyone done anything like that?

Dann Corbit
Posts: 11622
Joined: Wed Mar 08, 2006 7:57 pm
Location: Redmond, WA USA
Contact:

Re: Random playout vs evaluation

Post by Dann Corbit » Fri May 15, 2015 6:55 pm

Monte Carlo search is used all the time in go.
I don't know if anyone has tried it in chess.
https://chessprogramming.wikispaces.com ... ree+Search

Here is a github monte carlo tree search:
https://github.com/lteacy/mcts-cpp

AlvaroBegue
Posts: 925
Joined: Tue Mar 09, 2010 2:46 pm
Location: New York
Full name: Álvaro Begué (RuyDos)

Re: Random playout vs evaluation

Post by AlvaroBegue » Fri May 15, 2015 7:33 pm

I don't expect Monte-Carlo simulations would work very well in chess. Certainly not when giving every move the same probability ("light playouts", in MCTS lingo).

If you write something a bit smarter, so you don't drop pieces for no reason, perhaps you get some reasonable results. I would be curious to see a MCTS chess engine that uses depth-1 search during the playouts.

Daniel Shawul
Posts: 4050
Joined: Tue Mar 14, 2006 10:34 am
Location: Ethiopia
Contact:

Re: Random playout vs evaluation

Post by Daniel Shawul » Sat May 16, 2015 1:13 pm

AlvaroBegue wrote:I don't expect Monte-Carlo simulations would work very well in chess. Certainly not when giving every move the same probability ("light playouts", in MCTS lingo).

If you write something a bit smarter, so you don't drop pieces for no reason, perhaps you get some reasonable results. I would be curious to see a MCTS chess engine that uses depth-1 search during the playouts.
I tried this a while ago for chess and it sucked. MCTS works awesomely for checkers though because there captures are forced unlike in chess. And that seems to make a big difference more than anything else to the accuracy of the random playout. Nebiyu has MCTS mode for playing many types of games if anyone is interested.

flok

Re: Random playout vs evaluation

Post by flok » Sun May 17, 2015 12:31 pm

I tried it once, in a distributed version (+/- 120 computers).
Failed :-)

Post Reply