*First release* Giraffe, a new engine based on deep learning

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

Moderators: hgm, Harvey Williamson, bob

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Henk
Posts: 5799
Joined: Mon May 27, 2013 8:31 am

Re: *First release* Giraffe, a new engine based on deep lear

Post by Henk » Thu Jul 09, 2015 4:17 pm

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.
So it will be a competition between better evaluation versus an extra ply.

matthewlai
Posts: 791
Joined: Sun Aug 03, 2014 2:48 am
Location: London, UK
Contact:

Re: *First release* Giraffe, a new engine based on deep lear

Post by matthewlai » Thu Jul 09, 2015 4:20 pm

Henk wrote:
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.
So it will be a competition between better evaluation versus an extra ply.
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.
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.

Henk
Posts: 5799
Joined: Mon May 27, 2013 8:31 am

Re: *First release* Giraffe, a new engine based on deep lear

Post by Henk » Thu Jul 09, 2015 4:23 pm

matthewlai wrote:
Henk wrote:
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.
So it will be a competition between better evaluation versus an extra ply.
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.
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
Posts: 791
Joined: Sun Aug 03, 2014 2:48 am
Location: London, UK
Contact:

Re: *First release* Giraffe, a new engine based on deep lear

Post by matthewlai » Thu Jul 09, 2015 4:28 pm

Henk wrote:
matthewlai wrote:
Henk wrote:
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.
So it will be a competition between better evaluation versus an extra ply.
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.
Perhaps also or only use those neural networks for move ordering. But not to close to the leaves otherwise you get another computational bottleneck.
That's definitely something to think about.
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.

mvk
Posts: 589
Joined: Tue Jun 04, 2013 8:15 pm

Re: *First release* Giraffe, a new engine based on deep lear

Post by mvk » Thu Jul 09, 2015 5:08 pm

matthewlai wrote:the scores won't be comparable to other engine's scores, since it's probabilistic.
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.

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]

matthewlai
Posts: 791
Joined: Sun Aug 03, 2014 2:48 am
Location: London, UK
Contact:

Re: *First release* Giraffe, a new engine based on deep lear

Post by matthewlai » Thu Jul 09, 2015 5:27 pm

mvk wrote:
matthewlai wrote:the scores won't be comparable to other engine's scores, since it's probabilistic.
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.

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...
Indeed. The reason I don't really want to do that is not technical.

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.

modolief
Posts: 35
Joined: Tue Apr 30, 2013 4:29 pm

Re: *First release* Giraffe, a new engine based on deep lear

Post by modolief » Fri Jul 10, 2015 5:27 pm

Do you need compute resources for testing?

matthewlai
Posts: 791
Joined: Sun Aug 03, 2014 2:48 am
Location: London, UK
Contact:

Re: *First release* Giraffe, a new engine based on deep lear

Post by matthewlai » Fri Jul 10, 2015 10:26 pm

modolief wrote:Do you need compute resources for testing?
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.

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.

styx
Posts: 338
Joined: Tue Mar 13, 2012 8:59 pm
Location: Germany

Re: *First release* Giraffe, a new engine based on deep lear

Post by styx » Sat Jul 11, 2015 12:34 pm

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!

matthewlai
Posts: 791
Joined: Sun Aug 03, 2014 2:48 am
Location: London, UK
Contact:

Re: *First release* Giraffe, a new engine based on deep lear

Post by matthewlai » Sat Jul 11, 2015 12:37 pm

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!
Thanks for trying it out!

Isn't it nice when we can win against an engine with no handicap :D.
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.

Post Reply