MMTO for evaluation learning

Discussion of chess software programming and technical issues.

Moderators: hgm, Harvey Williamson, bob

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Post Reply
jdart
Posts: 3720
Joined: Fri Mar 10, 2006 4:23 am
Location: http://www.arasanchess.org

MMTO for evaluation learning

Post by jdart » Sun Jan 25, 2015 3:11 pm

This algorithm (MMTO), or a variant of it, is now used by all the top Shogi programs. For an objective function, it uses agreement between low-depth search results and game moves actually made by strong players.

There is some brief discussion of its application to chess and an experiment using Crafty in this paper:

https://www.jair.org/media/4217/live-4217-7792-jair.pdf

(see p. 555).

--Jon

jdart
Posts: 3720
Joined: Fri Mar 10, 2006 4:23 am
Location: http://www.arasanchess.org

Re: MMTO for evaluation learning

Post by jdart » Sun Jan 25, 2015 4:22 pm

Here is a less technical intro PPT, explains the error function better:

http://www.logos.ic.i.u-tokyo.ac.jp/~mi ... ummer.pptx

Ferdy
Posts: 3816
Joined: Sun Aug 10, 2008 1:15 pm
Location: Philippines

Re: MMTO for evaluation learning

Post by Ferdy » Sun Jan 25, 2015 9:34 pm

Thanks Jon, that is interesting. Something that it combines minimizing the error in score and at the same time guiding the score to choose the best move considered in a position.

wgarvin
Posts: 838
Joined: Thu Jul 05, 2007 3:03 pm
Location: British Columbia, Canada

Re: MMTO for evaluation learning

Post by wgarvin » Mon Jan 26, 2015 2:35 am

Weird that they mention KnightCap and the temporal-difference learning experiments that were done (Jonathan Baxter and Andrew Tridgell) but give a date of 2000 for some reason... I think they had published papers about it as early as 1998 (I was in university then and remember reading them).

This one for example: http://citeseerx.ist.psu.edu/viewdoc/su ... .1.36.7885

Anyways, it looks interesting, thanks for the links!

Post Reply