So it will be a competition between better evaluation versus an extra ply.matthewlai wrote: Yes, it's slower. The advantage is that it may learn things beyond humans' imaginations, and for me that's a cause worth pursuing.
*First release* Giraffe, a new engine based on deep learning
Moderators: hgm, Rebel, chrisw
-
- Posts: 7220
- Joined: Mon May 27, 2013 10:31 am
Re: *First release* Giraffe, a new engine based on deep lear
-
- Posts: 793
- Joined: Sun Aug 03, 2014 4:48 am
- Location: London, UK
Re: *First release* Giraffe, a new engine based on deep lear
I will also be working on tree-shaping later, also using neural networks. Ideally that would make the search more selective, and be able to reach deeper.Henk wrote:So it will be a competition between better evaluation versus an extra ply.matthewlai wrote: Yes, it's slower. The advantage is that it may learn things beyond humans' imaginations, and for me that's a cause worth pursuing.
Disclosure: I work for DeepMind on the AlphaZero project, but everything I say here is personal opinion and does not reflect the views of DeepMind / Alphabet.
-
- Posts: 7220
- Joined: Mon May 27, 2013 10:31 am
Re: *First release* Giraffe, a new engine based on deep lear
Perhaps also or only use those neural networks for move ordering. But not to close to the leaves otherwise you get another computational bottleneck.matthewlai wrote:I will also be working on tree-shaping later, also using neural networks. Ideally that would make the search more selective, and be able to reach deeper.Henk wrote:So it will be a competition between better evaluation versus an extra ply.matthewlai wrote: Yes, it's slower. The advantage is that it may learn things beyond humans' imaginations, and for me that's a cause worth pursuing.
-
- Posts: 793
- Joined: Sun Aug 03, 2014 4:48 am
- Location: London, UK
Re: *First release* Giraffe, a new engine based on deep lear
That's definitely something to think about.Henk wrote:Perhaps also or only use those neural networks for move ordering. But not to close to the leaves otherwise you get another computational bottleneck.matthewlai wrote:I will also be working on tree-shaping later, also using neural networks. Ideally that would make the search more selective, and be able to reach deeper.Henk wrote:So it will be a competition between better evaluation versus an extra ply.matthewlai wrote: Yes, it's slower. The advantage is that it may learn things beyond humans' imaginations, and for me that's a cause worth pursuing.
Disclosure: I work for DeepMind on the AlphaZero project, but everything I say here is personal opinion and does not reflect the views of DeepMind / Alphabet.
-
- Posts: 589
- Joined: Tue Jun 04, 2013 10:15 pm
Re: *First release* Giraffe, a new engine based on deep lear
You can map probabilistic scores to pawn scores with the inverse of the logistic function: 4*log10(p/(1-p)), if p is between 0 and 1.matthewlai wrote:the scores won't be comparable to other engine's scores, since it's probabilistic.
And, coincidently, if your probabilistic score is the output of a neuron, and if the step function of that neuron is the logistic function, then you might simply report the sum of that neuron's inputs...
[Account deleted]
-
- Posts: 793
- Joined: Sun Aug 03, 2014 4:48 am
- Location: London, UK
Re: *First release* Giraffe, a new engine based on deep lear
Indeed. The reason I don't really want to do that is not technical.mvk wrote:You can map probabilistic scores to pawn scores with the inverse of the logistic function: 4*log10(p/(1-p)), if p is between 0 and 1.matthewlai wrote:the scores won't be comparable to other engine's scores, since it's probabilistic.
And, coincidently, if your probabilistic score is the output of a neuron, and if the step function of that neuron is the logistic function, then you might simply report the sum of that neuron's inputs...
I want to use probabilistic scores internally in my program because I believe it is more "correct". I can just change the score in the printing function, but as I need to look at those scores a lot during testing, it could become confusing for myself.
That said, I will probably do it anyways for the next update.
The activation function of the output neuron is hyperbolic tangent.
Disclosure: I work for DeepMind on the AlphaZero project, but everything I say here is personal opinion and does not reflect the views of DeepMind / Alphabet.
-
- Posts: 44
- Joined: Tue Apr 30, 2013 6:29 pm
Re: *First release* Giraffe, a new engine based on deep lear
Do you need compute resources for testing?
-
- Posts: 793
- Joined: Sun Aug 03, 2014 4:48 am
- Location: London, UK
Re: *First release* Giraffe, a new engine based on deep lear
I may in the future, when I want to do more testing with other engines. I don't think it's quite at that stage yet.modolief wrote:Do you need compute resources for testing?
I have been using a computing cluster for testing and training, but the way it's set up means it may be difficult to actually run other people's engines (and also the cluster is obviously Linux-only).
Disclosure: I work for DeepMind on the AlphaZero project, but everything I say here is personal opinion and does not reflect the views of DeepMind / Alphabet.
-
- Posts: 338
- Joined: Tue Mar 13, 2012 9:59 pm
- Location: Germany
Re: *First release* Giraffe, a new engine based on deep lear
very interesting engine! it managed to win some fast games against me
somehow fun to play for a patzer like me
great achievement, good work!
somehow fun to play for a patzer like me
great achievement, good work!
-
- Posts: 793
- Joined: Sun Aug 03, 2014 4:48 am
- Location: London, UK
Re: *First release* Giraffe, a new engine based on deep lear
Thanks for trying it out!styx wrote:very interesting engine! it managed to win some fast games against me
somehow fun to play for a patzer like me
great achievement, good work!
Isn't it nice when we can win against an engine with no handicap .
Disclosure: I work for DeepMind on the AlphaZero project, but everything I say here is personal opinion and does not reflect the views of DeepMind / Alphabet.