Re: [Computer-go] Move Evaluation in Go Using Deep Convolutional Neural Networks
Hi Aja, Thanks for a game and report. I saw sgf, CNN can play ko fight. great. our best CNN is about 220 to 310 Elo stronger which is consistent Deeper network and rich info makes +300 Elo? impressive. Aja, if your CNN+MCTS use Erica's playout, how strong will it be? I think it will be contender for strongest program. I also wonder Fuego could release latest version with 1.2, and use odd number 1.3.x for development. Regards, Hiroshi Yamashita ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Move Evaluation in Go Using Deep Convolutional Neural Networks
Hiroshi Yamashita: 37E4294EAD9142EA84D1031F3E1E9C7C@x60: Hi Aja, Thanks for a game and report. I saw sgf, CNN can play ko fight. great. our best CNN is about 220 to 310 Elo stronger which is consistent Deeper network and rich info makes +300 Elo? impressive. Aja, if your CNN+MCTS use Erica's playout, how strong will it be? I think it will be contender for strongest program. The playing strength of an MCTS program is dominated by the correctness of the simulations, especially of LD. Prior knowledge helps a little. David pointed out after the first Densei-sen (almost three years ago): All mcts programs have trouble with the positions near the end. The group in the center has miai for two eyes. Same for the group at the top. The upper left side group has one big eye shape. For all three groups the playouts sometimes kill them. The black stones are pretty solid, so the playouts let them survivie. SO even at the end, zen has 50% win rate, MFGO has 60%, and pache has 70% win rate for blasck. Without improving the correctness of the simulations, MCTS programs can't go up to next stage. Hideki Hideki -- Hideki Kato mailto:hideki_ka...@ybb.ne.jp ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Move Evaluation in Go Using Deep Convolutional Neural Networks
I thought that any layers beyond 3 were irrelevant. Probably I'm subsuming your nn into what I learned about nn's and didn't read anything carefully enough. Can you help correct me? s. On Dec 23, 2014 6:47 AM, Aja Huang ajahu...@google.com wrote: On Mon, Dec 22, 2014 at 12:38 PM, David Silver davidstarsil...@gmail.com wrote: we'll evaluate against Fuego 1.1 and post the results. I quickly tested our 12-layer CNN against Fuego 1.1 with 5 secs and 10 secs per move, 2 threads. The hardware is Intel(R) Xeon(R) CPU E5-2687W 0 @ 3.10GHz. 5 secs per move 12-layer CNN scored 55.8% ±5.4 10 secs per move 12-layer CNN scored 32.9% ±3.8 Fuego1.1 is clearly much weaker than the latest svn release. And interestingly, the network is actually as strong as Fuego 1.1 with 5 secs per move. Since Clark and Storkey's CNN scored 12% against Fuego 1.1 running on a weaker hardware, our best CNN is about 220 to 310 Elo stronger which is consistent to the results against GnuGo. Regards, Aja ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Move Evaluation in Go Using Deep Convolutional Neural Networks
A 3-layer network (input, hidden, output) is sufficient to be a universal function approximator, so from a theoretical perspective only 3 layers are necessary. But the gap between theoretical and practical is quite large. The CNN architecture builds in translation invariance and sensitivity to local phenomena. That gives it a big advantage (on a per distinct weight basis) over the flat architecture. Additionally, the input layers of these CNN designs are very important. Compared to a stone-by-stone representation, the use of high level concepts in the input layer allows the network to devote its capacity to advanced concepts rather than synthesizing basic concepts. From: Computer-go [mailto:computer-go-boun...@computer-go.org] On Behalf Of uurtamo . Sent: Tuesday, December 23, 2014 7:34 PM To: computer-go Subject: Re: [Computer-go] Move Evaluation in Go Using Deep Convolutional Neural Networks I thought that any layers beyond 3 were irrelevant. Probably I'm subsuming your nn into what I learned about nn's and didn't read anything carefully enough. Can you help correct me? s. On Dec 23, 2014 6:47 AM, Aja Huang ajahu...@google.com mailto:ajahu...@google.com wrote: On Mon, Dec 22, 2014 at 12:38 PM, David Silver davidstarsil...@gmail.com mailto:davidstarsil...@gmail.com wrote: we'll evaluate against Fuego 1.1 and post the results. I quickly tested our 12-layer CNN against Fuego 1.1 with 5 secs and 10 secs per move, 2 threads. The hardware is Intel(R) Xeon(R) CPU E5-2687W 0 @ 3.10GHz. 5 secs per move 12-layer CNN scored 55.8% ±5.4 10 secs per move 12-layer CNN scored 32.9% ±3.8 Fuego1.1 is clearly much weaker than the latest svn release. And interestingly, the network is actually as strong as Fuego 1.1 with 5 secs per move. Since Clark and Storkey's CNN scored 12% against Fuego 1.1 running on a weaker hardware, our best CNN is about 220 to 310 Elo stronger which is consistent to the results against GnuGo. Regards, Aja ___ Computer-go mailing list Computer-go@computer-go.org mailto:Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Move Evaluation in Go Using Deep Convolutional Neural Networks
Whilst its technically true that you can use an nn with one hidden layer to learn the same function as a deeper net, you might need a combinatorally large number of nodes :-) scaling learning algorithms towards ai, by bengio and lecunn, 2007, makes a convincing case along these lines. ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] Fuego 1.1 vs current Fuego
Hello Hiroshi, we want to release a version 2.0. There is still some clean-up work to do for a release and progress is slow. But there is progress :) https://sourceforge.net/p/fuego/tickets/ https://sourceforge.net/p/fuego/tickets/ Martin I also wonder Fuego could release latest version with 1.2, and use odd number 1.3.x for development. ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
[Computer-go] Last move info as features
From: Stefan Kaitschick stefan.kaitsch...@hamburg.de mailto:stefan.kaitsch...@hamburg.de ... Last move info is a strange beast, isn't it? I mean, except for ko captures, it doesn't really add information to the position. The correct prediction rate is such an obvious metric, but maybe prediction shouldn't be improved at any price. To a certain degree, last move info is a kind of self-delusion. A predictor that does well without it should be a lot more robust, even if the percentages are poorer. My view is that what we really want to compare is temperature in terms of combinatorial game theory, in other words the urgency of playing locally. In Go, often the most urgent play remains within the same region for a while. Then things cool down, and the most urgent play moves elsewhere. So statistically, most of the urgent moves on the board are local replies. Lacking a direct measure of temperature/urgency, we use “reply locally when urgent-looking features exist” as our best cheap approximation. I have often wondered if a predictor could be trained to just answer this question “reply locally or not?” One problem of course is that sometimes the answer depends on the whole rest of the board. But often it does not. As a concrete experiment, we could define “answer locally” e.g. as “within cfg distance 4”, and then play the guessing game - how often do professionals and programs/predictors agree about answering locally? My guess would be quite often, maybe over 80%. The question is whether it would do any good. If you look at playout policies, they also struggle with the same issue of balancing local and nonlocal play in a meaningful way. Martin___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go