Forum Archive :
Programming
Bill Taylor asks:
> Some while ago, it was announced that a computer had beaten the then
> world backgammon champion. I believe the alleged "match" took place
> straight after the champs had been held.
>
> However, I also dimly recall reading that it was barely a match at all;
> in that it was only about 7 (?) games long. Hardly decisive, I'm sure
> you'll agree!
Actually it was only 5 games, but the program won 7-1 !!
> So anyway; can someone here shed light on all this. For instance...
>
> 1. When was it.
It was in June 1979 (yes! THAT long ago!)
> 2. Who was the world champ involved.
It was Luigi Villa from Italy
> 3. Who/what were the machine/program/programmers.
Hans Berliner, an excellent chess player (world correspondence-chess
champion). He was supported by the famous Paul Magriel, who helped
refining some algorithms and improving the program's overall performance.
> 4. What really happened.
Villa got backgammon world champion 1979 in Monte Carlo and played against
Berliners BKG 9.8 program after that.
The scores were the following:
game BKG 9.8 Villa
1 2 - 0
2 1 - 0
3 2 - 0
4 0 - 1
5 2 - 0
Berliner won $5000 and wrote afterwards:
Villa, who only a day earlier had reached the summit of his backgammon
career in winning the world title, was disconsolate. I told him that I
was sorry it had happened and that we both knew he was really the
better player.
Several weeks later I analyzed the games in some detail [...]. There was
no doubt that BKG 9.8 played well, but down the line Villa played better.
He made the technically correct plays almost all the time, whereas the
program did not make the best play in eight out of 73 nonforced
situations. Only one of the mistakes, however gave the program any
trouble. An expert would not have made most of the errors, but the could
be exploited only a small percent of the time.
That's wonderful in backgammon: Even a computer program can win against
the world champion, although it plays worse :-)
> 5. Have any similar matchings occurred since.
I never heard about a world champion who dared playing against a computer
in the public after Villa :-)
> 6. Just how good ARE machines at backgammon now? I know at chess they've
> become close to grandmaster level. But at go, for instance, they're still
> a biiiiiig dissappointment (brute-force is ineffective here).
All backgammon programs I played are miserable (and I'm not a champion at
all). BKG 9.8 was never available on the market as far as I know and
Berliner hasn't publshed anything about it since his article (it was in the
Scientific American if I recall it right - Don't ask me for the year). If
he did: please let me know! In his article he also described how his
program worked. There are really some very nice ideas in the program.
Andreas
--
Email: marvin@ouzo.rog.rwth-aachen.de
marvin@pool.informatik.rwth-aachen.de
Mail: Andreas Schneider, Kapuzinergraben 8, 5100 Aachen, Germany
Phone: +49-241-35562
|
|
|
|
Programming
- 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)
- Blockading feature (Sam Pottle+, Feb 1999)
- 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)
- 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)
- 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)
- Variance reduction of rollouts (Michael J. Zehr+, Aug 1998)
- 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)
From GammOnLine
Long message
Recommended reading
Recent addition
|
| |
|