Forum Archive : Programming

Training for different gammon values

From:   Gerry Tesauro
Date:   2 February 1996
Subject:   Re: Maybe Mloner knows something
Google:   4erluu$

Bob Koca wrote:
>  On a related note is this a good idea for neural nets?:
> Train separately for double match point and for one way gammon
> positions. Seems like it should not be too much extra effort
> since previously obtained weights for money play could be
> used as the starting point and difference will not be huge.
> I think the gain in playing ability near the end of a match
> could be substantial.

This seems like a potentially good idea, because the regular
neural net trained for money play will give you win and gammon
estimates *assuming money-type plays are made* by both sides.
Thus it's conceivable that if we trained a win estimator
under conditions of double match point where gammons don't
count, then such a win estimator might be different from
the win estimator that comes out of money-condition training.

However, this appears not to be the case empirically.  I once
trained a double match point version of TD-Gammon, and compared
it to regular TD with the gammon estimates turned off.
There seemed to be no measurable advantage in playing ability
of the DMP version, at least at the 1-ply level.  One-way
gammons I don't know about, but I'd have to guess that you
get reasonably correct play just by taking the regular
money-play net and re-weighting the win and gammon outputs
appropriately.  Perhaps Fredrik or Harald could contribute
further to the discussion.

-- Gerry Tesauro (
Did you find the information in this article useful?          

Do you have any comments you'd like to add?     



Adjusting to a weaker opponent  (Brian Sheppard, July 1997) 
Anticomputer positions  (Bill Taylor+, June 1998) 
BKG 9.8 vs. Villa  (Raccoon+, Aug 2006) 
BKG 9.8 vs. Villa  (Andreas Schneider, June 1992) 
BKG beats world champion  (Marty Storer, Sept 1991) 
Backgames  (David Montgomery+, June 1998)  [Long message]
Blockading feature  (Sam Pottle+, Feb 1999)  [Long message]
Board encoding for neural network  (Brian Sheppard, Feb 1997) 
Bot weaknesses  (Douglas Zare, Mar 2003) 
Building and training a neural-net player  (Brian Sheppard, Aug 1998) 
How to count plies?  (Chuck Bower+, Jan 2004)  [GammOnLine forum]
How to count plies?  (tanglebear+, Mar 2003) 
Ideas for improving computer play  (David Montgomery, Feb 1994) 
Ideas on computer players  (Brian Sheppard, Feb 1997) 
Introduction  (Gareth McCaughan, Oct 1994) 
Measuring Difficulty  (John Robson+, Feb 2005)  [GammOnLine forum]
Methods of encoding positions  (Gary Wong, Jan 2001) 
N-ply algorithm  (eXtreme Gammon, Jan 2011) 
Neural net questions  (Brian Sheppard, Mar 1999) 
Pruning the list of moves  (David Montgomery+, Feb 1994) 
Search in Trees with Chance Nodes  (Thomas Hauk, Feb 2004) 
Source code  (Gary Wong, Dec 1999) 
TD-Gammon vs. Robertie  (David Escoffery, June 1992) 
Training for different gammon values  (Gerry Tesauro, Feb 1996) 
Training neural nets  (Walter Trice, Nov 2000) 
Variance reduction in races  (David Montgomery+, Dec 1998)  [Long message]
Variance reduction of rollouts  (Michael J. Zehr+, Aug 1998)  [Long message]
Variance reduction of rollouts  (Jim Williams, June 1997) 
What is a "neural net"?  (Gary Wong, Oct 1998) 
Writing a backgammon program  (Gary Wong, Jan 1999) 

[GammOnLine forum]  From GammOnLine       [Long message]  Long message       [Recommended reading]  Recommended reading       [Recent addition]  Recent addition

  Book Suggestions
Computer Dice
Cube Handling
Cube Handling in Races
Extreme Gammon
Fun and frustration
GNU Backgammon
Luck versus Skill
Magazines & E-zines
Match Archives
Match Equities
Match Play
Match Play at 2-away/2-away
Opening Rolls
Pip Counting
Play Sites
Probability and Statistics
Source Code
Strategy--Bearing Off
Strategy--Checker play


Return to:  Backgammon Galore : Forum Archive Main Page