Human Shogi players still stronger than computer............

Discussion of anything and everything relating to chess playing software and machines.

Moderators: bob, hgm, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
User avatar
George
Posts: 682
Joined: Sun Oct 13, 2013 11:44 pm

Re: Human Shogi players still stronger than computer........

Post by George » Sat Apr 18, 2015 4:43 pm

myfish wrote:Shogi programs are getting stronger. No doubt for the average player, they are VERY formidable and myself I find many engines simply too strong. I have, rarely beaten some of the more common ones but it is, occasional rather than the norm.

Shogi is hugely technical and strategical even by comparison to an already tough and hugely strategical western chess. The permutations are almost, whist not infinite, they seem that way.

Just the other day, I played a human <> human game where it was deadlock for a long time and the focus changed from being a king game, to that of trying to get a piece in hand, held at the time, by your opponent and in a strong position itself, to start the game to mate with the king.

And, I think that is quite common where players play strategies that in the long term will or may rely on a piece that is in your opponents possession.

Maybe it's this depth, that is separating the computer shogi engines really looking as strong as the professional. Like I say, ordinary players have a formidable opponent in many top shogi engines. Too strong in most cases for kyu players but then, there seems to be this almost spiritual leap to 'knowing' and 'seeing' the engines haven't grasped yet.

But, I think in time, IF the pro's play the newer engines more and more and we see the results, this gap may narrow in the next decade.
According to the world Japanese Shogi champion Mr. Habu who is my friend via Facebook, he already think that the best Shogi program in a math of 5 games could beat him since they are as strong as him, computer can always beat his human stamina in the long run, but in a short match of only 3 games he thinks that he still has the edge, but after the 3rd day he would be exhausted trying to calculate all the variations and possibilities. Remember that a Shogi match could go at long as 9 hours a side plus extra time in the prestigious Meijin title match, and that can cause severe stress to any human.

As you can observe in 2013 GPS-Shogi beat one of Japan strongest Shogi player
==>
http://www.japantimes.co.jp/news/2013/0 ... TKEpCFViko

Mr. Habu is also Japan best Western Chess Player with hardly any training, he is a natural genius ==> https://www.youtube.com/watch?v=QUomyItgWG4

Here is Another interesting game between Mr. Kasparov vs Mr. Habu ==>
https://www.youtube.com/watch?v=t2da69omY1M

Dirt
Posts: 2851
Joined: Wed Mar 08, 2006 9:01 pm
Location: Irvine, CA, USA

Re: Human Shogi players still stronger than computer........

Post by Dirt » Sat Apr 18, 2015 6:43 pm

George wrote:The computer system used for Shogi competition vs human in Japan are much much much stronger than any computer used by stockfish so far, but the complexity of Shogi compared to Chess is much much harder since the captured pieces can be used by the opponent and placed back into the board immediately and when you do NOT see a checkmate the captured pieces will come out of nowhere and be placed where checkmate is possible, also to make it harder Shogi is played on a 9x9 board instead of 8x8 like chess is.
I don't think the drops make Shogi particularly hard for computers relative to humans, rather it is more the weak pieces. You see this in other games with weak pieces, like Arimaa and Go.
Deasil is the right way to go.

User avatar
mschribr
Posts: 223
Joined: Thu Mar 09, 2006 3:23 am
Location: new york ny usa

Re: Human Shogi players still stronger than computer........

Post by mschribr » Mon Apr 20, 2015 7:05 pm

George wrote:After 15 years since Deeper Blue beat Kasparov in 1997, in Japan Shogi programers are trying hard to beat Shogi professionals, and they are getting closer but it will be a few years until a computer can beat the best Shogi player and World Champion.
First, Shogi does not belong in the chess forum. It should be on a shogi forum or at least on the Chess Thinkers Forum.
These five shogi games do not show it will be a few years until a computer can beat the best Shogi player, because the Shogi programmers were restricted in which computer they could use. They could only use a single CPU computer. These five games do show that some shogi programs are so strong that the programs beat strong players even though the programs were running on a single CPU. In addition, the players practiced with programs for 5 months. Also, the programmers were not allowed to change their programs for those 5 months. Remember that in 2013 a computer beat the 20th highest ranked player. According to the floodgate server the best programs improved by 368 points from 2013 to 2015. A 32 CPU cluster would destroy the best Shogi player.
Have you heard a computer won the arimaa challenge even though the rules favored the human players?
Mark

User avatar
George
Posts: 682
Joined: Sun Oct 13, 2013 11:44 pm

Re: Human Shogi players still stronger than computer........

Post by George » Tue Apr 21, 2015 2:08 am

mschribr wrote:
George wrote:After 15 years since Deeper Blue beat Kasparov in 1997, in Japan Shogi programers are trying hard to beat Shogi professionals, and they are getting closer but it will be a few years until a computer can beat the best Shogi player and World Champion.
First, Shogi does not belong in the chess forum. It should be on a shogi forum or at least on the Chess Thinkers Forum.

Sorry for talking about Shogi on the wrong location, if it make you more comfortable you can continue here in the Thinkers Chess Forum Chess thinkers Forum "A place for chess players to speak about what interests them: chess or otherwise". ==>
http://talkchess.com/forum/viewtopic.ph ... highlight=
These five shogi games do not show it will be a few years until a computer can beat the best Shogi player, because the Shogi programmers were restricted in which computer they could use. They could only use a single CPU computer. These five games do show that some shogi programs are so strong that the programs beat strong players even though the programs were running on a single CPU. In addition, the players practiced with programs for 5 months. Also, the programmers were not allowed to change their programs for those 5 months. Remember that in 2013 a computer beat the 20th highest ranked player. According to the floodgate server the best programs improved by 368 points from 2013 to 2015. A 32 CPU cluster would destroy the best Shogi player.
Have you heard a computer won the arimaa challenge even though the rules favored the human players?
Mark

User avatar
George
Posts: 682
Joined: Sun Oct 13, 2013 11:44 pm

Re: Human Shogi players still stronger than computer........

Post by George » Wed Apr 22, 2015 11:58 pm

lkaufman wrote:Actually Don and I did make a "Komodoshogi" and it played on the server a bit. But we only spent two weeks on it so it was only about 2000 rated there. A Japanese backer expressed interest but didn't follow thru, and Don could not afford to work for free. It's pretty hard for a non-Japanese to get into the Japanese shogi market. The hard part of the program is indeed the eval, which is far more crucial to elo in shogi than in chess.
Larry back in 2011 you thought that a Shogi program was going to be strong enough to beat the current champion Mr. Habu, but even on a single processor like your Komodo or Stockfish can beat Mr. Carlsen, nowaday or a few months back the best Shogi program were unable to beat the human team.

Larry this was your quote back in 2011 "By lkaufman Date 2011-03-07 03:05 Shogi is certainly more tactically complex than chess, but the evaluation of positions is even more difficult than in chess. It just happens that in chess material is overwhelmingly important, it takes a lot of compensation just for one pawn, and having adequate positional compensation for two pawns or the equivalent is quite rare, rare enough that a program that assumed it was impossible would probably only give up a handful of Elo points. But in shogi it is far easier to have compensation for material, and the tradeoff is very hard to judge. It is quite difficult to teach a program to make this judgement as well as a top shogi pro can make it. Just adding up a bunch of positional features the way chess programs do just won't work. If you want to say that this makes shogi a more positional game than chess, well that's a matter of definition. I would say that it's a game where proper "positional" play is relatively more important than in chess, but not necessarily more complex. A great deal of effort has already gone into making strong shogi programs, ok less than in chess but still quite a lot, and the problems haven't been fully solved, though the rate of progress is impressive. I'll predict 3 years until a program is equal to the Champ."

lkaufman
Posts: 3741
Joined: Sun Jan 10, 2010 5:15 am
Location: Maryland USA
Contact:

Re: Human Shogi players still stronger than computer........

Post by lkaufman » Thu Apr 23, 2015 12:06 am

George wrote:
lkaufman wrote:Actually Don and I did make a "Komodoshogi" and it played on the server a bit. But we only spent two weeks on it so it was only about 2000 rated there. A Japanese backer expressed interest but didn't follow thru, and Don could not afford to work for free. It's pretty hard for a non-Japanese to get into the Japanese shogi market. The hard part of the program is indeed the eval, which is far more crucial to elo in shogi than in chess.
Larry back in 2011 you thought that a Shogi program was going to be strong enough to beat the current champion Mr. Habu, but even on a single processor like your Komodo or Stockfish can beat Mr. Carlsen, nowaday or a few months back the best Shogi program were unable to beat the human team.

Larry this was your quote back in 2011 "By lkaufman (*****) Date 2011-03-07 03:05 Shogi is certainly more tactically complex than chess, but the evaluation of positions is even more difficult than in chess. It just happens that in chess material is overwhelmingly important, it takes a lot of compensation just for one pawn, and having adequate positional compensation for two pawns or the equivalent is quite rare, rare enough that a program that assumed it was impossible would probably only give up a handful of Elo points. But in shogi it is far easier to have compensation for material, and the tradeoff is very hard to judge. It is quite difficult to teach a program to make this judgement as well as a top shogi pro can make it. Just adding up a bunch of positional features the way chess programs do just won't work. If you want to say that this makes shogi a more positional game than chess, well that's a matter of definition. I would say that it's a game where proper "positional" play is relatively more important than in chess, but not necessarily more complex. A great deal of effort has already gone into making strong shogi programs, ok less than in chess but still quite a lot, and the problems haven't been fully solved, though the rate of progress is impressive. I'll predict 3 years until a program is equal to the Champ."
I think I was about right. It seems to me that in a fair match (no copy of code to player, no limit on cores or say a 32 core limit) the betting favorite today would be the best engine against Habu. Perhaps we'll never know.
Komodo rules!

Post Reply