Re: [Computer-go] MCTS + Neural Networks?
We used neural networks (learnt by DirectPolicySearch) for making MCTS compliant with problems with long horizons. But this was not for the game of Go. http://hal.inria.fr/hal-00759822/ MCTS has a problem for long horizons, and a Direct Policy Search with neural networks is a solution for learning a default playout. Best regards, Olivier 2013/5/2 Steven Clark steven.p.cl...@gmail.com Hello all- Has anyone successfully used neural nets to help guide MC playouts? Has anyone used NN to learn patterns larger than 3x3? I'm working on a grad-school project, and discovered a few interesting things. After analyzing 10,000+ high-dan games from KGS, I find that more than 50% of the time, moves are played within a 5x5 window centered at the opponent's previous move (call this a tactical move, vs a strategic move). I used the FANN library to learn these 5x5 patterns, and found that the NN could predict tactical moves with ~27% accuracy (and with a 50% chance that the answer would be in the top 3 moves proposed by the NN). Is this old news? Are neural nets just too slow to be helpful to MC (reduce the playout rate too much?) Thoughts welcome. I will be up late finishing the report since it is due tomorrow ;) -Steven ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go -- = Olivier Teytaud, olivier.teyt...@inria.fr, TAO, LRI, UMR 8623(CNRS - Univ. Paris-Sud), bat 490 Univ. Paris-Sud F-91405 Orsay Cedex France http://www.slideshare.net/teytaud ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
[Computer-go] Fwd: MCTS + Neural Networks?
We used neural networks (learnt by DirectPolicySearch) for making MCTS compliant with problems with long horizons. But this was not for the game of Go. http://hal.inria.fr/hal-00759822/ MCTS has a problem for long horizons, and a Direct Policy Search with neural networks is a solution for learning a default playout. Best regards, Olivier 2013/5/2 Steven Clark steven.p.cl...@gmail.com Hello all- Has anyone successfully used neural nets to help guide MC playouts? Has anyone used NN to learn patterns larger than 3x3? I'm working on a grad-school project, and discovered a few interesting things. After analyzing 10,000+ high-dan games from KGS, I find that more than 50% of the time, moves are played within a 5x5 window centered at the opponent's previous move (call this a tactical move, vs a strategic move). I used the FANN library to learn these 5x5 patterns, and found that the NN could predict tactical moves with ~27% accuracy (and with a 50% chance that the answer would be in the top 3 moves proposed by the NN). Is this old news? Are neural nets just too slow to be helpful to MC (reduce the playout rate too much?) Thoughts welcome. I will be up late finishing the report since it is due tomorrow ;) -Steven ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go -- = Olivier Teytaud, olivier.teyt...@inria.fr, TAO, LRI, UMR 8623(CNRS - Univ. Paris-Sud), bat 490 Univ. Paris-Sud F-91405 Orsay Cedex France http://www.slideshare.net/teytaud -- = Olivier Teytaud, olivier.teyt...@inria.fr, TAO, LRI, UMR 8623(CNRS - Univ. Paris-Sud), bat 490 Univ. Paris-Sud F-91405 Orsay Cedex France http://www.slideshare.net/teytaud ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] MCTS + Neural Networks?
A recent comparison of move prediction systems for Go can be found here (IEEE CIG 2012 proceedings) http://geneura.ugr.es/cig2012/papers/paper82.pdf Simon Lucas From: computer-go-boun...@dvandva.org [mailto:computer-go-boun...@dvandva.org] On Behalf Of Steven Clark Sent: 02 May 2013 03:10 To: computer-go@dvandva.org Subject: Re: [Computer-go] MCTS + Neural Networks? Thanks for the link! Looks like a good paper -- I will read it more carefully shortly. Ignoring computational speed for a moment, is it a reasonable assumption that an algorithm that plays a NN-proposed tactical move 50% of the time, and a random move 50% of the time, should outperform an algorithm that plays a random move 100% of the time? So it's just a case of how many playouts do we lose by employing the NN (GPUs to the rescue?). For reference, I was using 25 input nodes, 25 output nodes, ~50 hidden nodes. I guess ultimately it comes down to make a bot and prove it :) -Steven On Wed, May 1, 2013 at 9:50 PM, George Dahl gd...@cs.toronto.edumailto:gd...@cs.toronto.edu wrote: I don't know if neural nets that predict moves have been helpful in any strong bots, but predicting expert moves with neural nets is certainly old news. See http://www.cs.utoronto.ca/~ilya/pubs/2008/go_paper.pdf There might be a place for artificial neural nets in a strong Go playing program, but it is an open question on how to incorporate neural nets well. Software like neurgo used a lot of expert features along with a neural net for global position evaluation and I tried (with very little success) to predict ownership of points on the board using a neural net. It is very hard to get neural nets to help a standard MCTS bot a lot because the neural net needs to be good at whatever it is supposed to be doing and still probably very fast to be useful. - George On Wed, May 1, 2013 at 9:42 PM, Steven Clark steven.p.cl...@gmail.commailto:steven.p.cl...@gmail.com wrote: Hello all- Has anyone successfully used neural nets to help guide MC playouts? Has anyone used NN to learn patterns larger than 3x3? I'm working on a grad-school project, and discovered a few interesting things. After analyzing 10,000+ high-dan games from KGS, I find that more than 50% of the time, moves are played within a 5x5 window centered at the opponent's previous move (call this a tactical move, vs a strategic move). I used the FANN library to learn these 5x5 patterns, and found that the NN could predict tactical moves with ~27% accuracy (and with a 50% chance that the answer would be in the top 3 moves proposed by the NN). Is this old news? Are neural nets just too slow to be helpful to MC (reduce the playout rate too much?) Thoughts welcome. I will be up late finishing the report since it is due tomorrow ;) -Steven ___ Computer-go mailing list Computer-go@dvandva.orgmailto:Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go ___ Computer-go mailing list Computer-go@dvandva.orgmailto:Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
[Computer-go] May KGS computer Go tournament - 19x19
The May KGS bot tournament will start at 16:00 UTC on the (European) afternoon of Sunday May 12th, and end at 22:00 UTC that evening. It will have 12 rounds, Swiss, with 19x19 boards. The time limits will be 14 minutes each, with fast Canadian overtime of 10 moves in 30 seconds. There are details at http://www.gokgs.com/tournInfo.jsp?id=807 . Please register by emailing me, with the words KGS Tournament Registration in the email title, at mapr...@gmail.com . Nick -- Nick Wedd n...@maproom.co.uk ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
[Computer-go] Computer Go at the European Go Congress
The 2013 European Go Congress will be held in Olsztyn, Poland, from July 27th to August 11th. On Friday August 2nd, the EGC Science Conference will be held there, see http://egc2013.go.art.pl/page/egc-science-conference . It will include three computer Go tournaments, see the final paragraph of http://egc2013.go.art.pl/page/go-tournaments I learned today, from the organiser Paweł Noga, that entrants for these tournaments will be required to use the computers provided. I do not yet know the specification of these computers, though I guess they will be running Windows. I also do not yet know whether programmers will be allowed to appoint someone who is at the Congress, to operate their programs for them. I have asked Paweł about both. I would like to hear from the owners of strong programs whether they are planning to enter. If no strong program will be competing, I will not want to travel to Poland for the event. If there will be at least two strong programs competing, I will travel to Poland, and will offer to operate competing programs for those who cannot be there themselves, if this is to be permitted. If I am to be in Poland for the Congress, the August KGS bot tournament will probably be on August 25th; if I am not going there, it will be on August 4th. Nick -- Nick Wedd n...@maproom.co.uk ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] Computer Go at the European Go Congress
I will enter if someone can run the program for me. Many faces only runs on Windows, but it would not be too hard to make a Linux version (no GUI, GTP control, for KGS). Have they said how the computers will communicate? KGS would be ideal, sine the operator would just run a script before the tournament then watch all the games. David -Original Message- From: computer-go-boun...@dvandva.org [mailto:computer-go- boun...@dvandva.org] On Behalf Of Nick Wedd Sent: Thursday, May 02, 2013 3:54 AM To: computer-go@dvandva.org Subject: [Computer-go] Computer Go at the European Go Congress The 2013 European Go Congress will be held in Olsztyn, Poland, from July 27th to August 11th. On Friday August 2nd, the EGC Science Conference will be held there, see http://egc2013.go.art.pl/page/egc-science- conference . It will include three computer Go tournaments, see the final paragraph of http://egc2013.go.art.pl/page/go-tournaments I learned today, from the organiser Pawel Noga, that entrants for these tournaments will be required to use the computers provided. I do not yet know the specification of these computers, though I guess they will be running Windows. I also do not yet know whether programmers will be allowed to appoint someone who is at the Congress, to operate their programs for them. I have asked Pawel about both. I would like to hear from the owners of strong programs whether they are planning to enter. If no strong program will be competing, I will not want to travel to Poland for the event. If there will be at least two strong programs competing, I will travel to Poland, and will offer to operate competing programs for those who cannot be there themselves, if this is to be permitted. If I am to be in Poland for the Congress, the August KGS bot tournament will probably be on August 25th; if I am not going there, it will be on August 4th. Nick -- Nick Wedd n...@maproom.co.uk ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] Computer Go at the European Go Congress
Hi! On Thu, May 02, 2013 at 11:53:57AM +0100, Nick Wedd wrote: I would like to hear from the owners of strong programs whether they are planning to enter. If no strong program will be competing, I will not want to travel to Poland for the event. If there will be at least two strong programs competing, I will travel to Poland, and will offer to operate competing programs for those who cannot be there themselves, if this is to be permitted. I am considering to travel to Poland, but I am not decided yet whether I will go. I'm waiting for some sort of official announcement of the tournament with more detailed information; from http://egc2013.go.art.pl/page/go-tournaments I've understood that we could install Linux on the competition computer. -- Petr Pasky Baudis For every complex problem there is an answer that is clear, simple, and wrong. -- H. L. Mencken ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go