Today I've done a little match between Satana and Giraffe. You can find the games on my website www.linformatica.com
The style of Giraffe is very interesting... and strange almost as those of satana
*First release* Giraffe, a new engine based on deep learning
Moderators: hgm, Rebel, chrisw
-
- Posts: 859
- Joined: Mon Aug 10, 2009 10:05 pm
- Location: Italy
- Full name: Stefano Gemma
-
- 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! The games look pretty interesting . One big problem with Giraffe right now is that she always wants to get the king out early. That's probably positions in endgames are trained faster (because they are close to the final reward), so the eval is biased for endgames, where the king should get out.stegemma wrote:Today I've done a little match between Satana and Giraffe. You can find the games on my website www.linformatica.com
The style of Giraffe is very interesting... and strange almost as those of satana
It also really likes to push pawns, probably for the same reason.
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: 859
- Joined: Mon Aug 10, 2009 10:05 pm
- Location: Italy
- Full name: Stefano Gemma
Re: *First release* Giraffe, a new engine based on deep lear
Satana has similar problem: it moves the queen too soon and rarely it play castling. But don't worry, even young players starts this way... so your neural network is working like a child's brain!matthewlai wrote:Thanks! The games look pretty interesting . One big problem with Giraffe right now is that she always wants to get the king out early. That's probably positions in endgames are trained faster (because they are close to the final reward), so the eval is biased for endgames, where the king should get out.stegemma wrote:Today I've done a little match between Satana and Giraffe. You can find the games on my website www.linformatica.com
The style of Giraffe is very interesting... and strange almost as those of satana
It also really likes to push pawns, probably for the same reason.
-
- 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 is always good to know .stegemma wrote:Satana has similar problem: it moves the queen too soon and rarely it play castling. But don't worry, even young players starts this way... so your neural network is working like a child's brain!matthewlai wrote:Thanks! The games look pretty interesting . One big problem with Giraffe right now is that she always wants to get the king out early. That's probably positions in endgames are trained faster (because they are close to the final reward), so the eval is biased for endgames, where the king should get out.stegemma wrote:Today I've done a little match between Satana and Giraffe. You can find the games on my website www.linformatica.com
The style of Giraffe is very interesting... and strange almost as those of satana
It also really likes to push pawns, probably for the same reason.
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: 1766
- Joined: Wed Jun 03, 2009 12:14 am
Re: *First release* Giraffe, a new engine based on deep lear
is it based on learning at an end user level, or is that simply how its developing on your end?
ie will it learn as i play it or run it against other engines?
ie will it learn as i play it or run it against other engines?
-
- Posts: 793
- Joined: Sun Aug 03, 2014 4:48 am
- Location: London, UK
Re: *First release* Giraffe, a new engine based on deep lear
It is learning on my end only. There's no learning through normal gameplay.yanquis1972 wrote:is it based on learning at an end user level, or is that simply how its developing on your end?
ie will it learn as i play it or run it against other engines?
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: 1766
- Joined: Wed Jun 03, 2009 12:14 am
Re: *First release* Giraffe, a new engine based on deep lear
thanks -- very cool to see someone taking a novel approach to engineering a chess program. the high level engines are great but i definitely think theres a 'market' for interesting and unique engines as well.
-
- Posts: 793
- Joined: Sun Aug 03, 2014 4:48 am
- Location: London, UK
Re: *First release* Giraffe, a new engine based on deep lear
Yeah it's a lot of fun .yanquis1972 wrote:thanks -- very cool to see someone taking a novel approach to engineering a chess program. the high level engines are great but i definitely think theres a 'market' for interesting and unique engines as well.
With conventional approaches I'm always out of ideas to try, and most of the ideas I do try do no better than existing approaches, which is to be expected since existing approaches have been fine-tuned for decades already.
With this new approach almost everything is still unexplored. I have a huge list of stuff I want to try, and quite a few of them have resulted in good improvements. I can't implement them fast enough!
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: 186
- Joined: Fri Oct 10, 2014 10:05 pm
- Location: Berkeley, CA
Re: *First release* Giraffe, a new engine based on deep lear
Very interesting project. I have two questions:
1. Why did you bootstrap from material-only eval? Doesn't the CCRL dump contain evals by the best engines? Why not train on those?
2. Are you familiar with the idea of model compression -- training a small network to mimic a larger one? E.g.
http://arxiv.org/abs/1312.6184
http://arxiv.org/abs/1503.02531
and do you think this could be used to speed up Giraffe's eval?
Cheers,
-Carl
1. Why did you bootstrap from material-only eval? Doesn't the CCRL dump contain evals by the best engines? Why not train on those?
2. Are you familiar with the idea of model compression -- training a small network to mimic a larger one? E.g.
http://arxiv.org/abs/1312.6184
http://arxiv.org/abs/1503.02531
and do you think this could be used to speed up Giraffe's eval?
Cheers,
-Carl
-
- Posts: 213
- Joined: Thu Dec 16, 2010 4:39 pm
Re: *First release* Giraffe, a new engine based on deep lear
One big problem with Giraffe right now is that she always wants to get the king out early. That's probably positions in endgames are trained faster (because they are close to the final reward), so the eval is biased for endgames, where the king should get out.
Could you not compensate for this training bias by providing more context parameters, actually modelling what we, as humans, understand as the 3 game phases in chess (opening, middlegame, endgame)?
Doing so, development moves (but not king move beside castling, lol) would be favored in opening and king centralization (a strong point in Giraffe's endgame play) reserved for endgame?
Per ardua ad astra