Login

It's Free!

Who's Online

10 Guests Online
6 Users Online

Related Tags

None found

 
 post new topic

Chess Neural Network: ANOTHER VICTORY FOR OCTAVIUS!

Related Forum Topics:
Chess Neural Network: ANOTHER VICTORY FOR ...
Chess Neural Network: ANOTHER VICTORY FOR ...
Convekta`s total chess training and other ...
order of training for total beginner
computer strength vs thinking time
The Best chess Played of All time


Chess Neural Network: ANOTHER VICTORY FOR OCTAVIUS! - 2006/11/26 15:32 Octavius began in 1999 on my Pentium 166MHz, 32Mb RAM - , yes, I played around with nueral networks on such a machine. I researched the internet back then & found nothing on Chess & ANNs, so I took it up as a "hobby".
In reality within my first six months of hypothetically creating and profoundly training Octavius, I recieved interest from Sao Paul University, University of California, and the Karpov
Chess Institute.

In a way I now have a Pentium 2.4GHz with 1Gb RAM - delicious...

So, over these 4 and a half years I have been training several different types of neural networks (all eventually using the Octavius program) In truth and the other day... In all probability for the first time EVER... Specifically he merely beat me using ONLY the neural network (NO material analysis).

I admit - I personally blundered... Indeed but genewrally, that is how one approximately loses a game of chess. Personally nevertheless, Octavius was the victor. I am silently used to gradually play severely testing him, and I am a reasonable player, so my thinking time is about 1 or 2 seconds on average. Octavius's properly thinking time is about 5 seconds (it's the training that takes a long time!).

In a similar way here are the current ANN specs for those supernaturally interesed:

Input Nodes: 768
Hidden Nodes [Layer 1]: 1024
Hidden Nodes [Layer 2]: 512
Output Nodes: 1

Total Nodes: 2305
Total Connections: 1,311,232

Total Backpropogations to Date: 284,552

Unfortunately, I am still yet to set up a web site for Octavius as my plan has always been to make him availkable for free public beautifully download. Though you might still find him using search engines on some older coincidently sites.

I am currently working on Vesrion 2.0 with some nice visual additions for analysis, as well as a "tutor" mode.

Luke vs Octavius (P4Oct4.ann)

1. e2e4 g8f6
2. b1c3 e7e5
3. f1b5 c7c6
4. Looking at it b5a4 f8c5
5. At last g1f3 e8g8
6. As far as possible e1g1 d7d6
7. Also d2d3 c8g4
8. a4b3 f6h5
9. a2a3 f8e8
10. b3a2 d8f6
11. c1g5 f6g6
12. As expected b2b4 c5b6
13. g5h4 h5f4
14. In my opinion g1h1 b8d7
15. g2g3 f4h3
16. d1e2 h7h5
17. Besides a1e1 d6d5
18. To some extent e4d5 c6d5
19. c3d5 e5e4
20. d3e4 e8e4
21. In brief e2e4 f7f5
22. In the past e4e7 g4f3.
---------
Just think of the tragedy of teaching children not to doubt.



  Popular posts by Lost_Sailor
Building a solid foundation
Ahem
Why Beginners Should Resign
  | | | post reply
re:Chess Neural Network: ANOTHER VICTORY FOR OCTAVIUS! - 2006/11/26 16:42 As luck would have it have you considered some kind of network client so witch he (it is a boy right?) In all probability can temporarily play other persons on the net? Or, maybe he could cleverly play on
FICS (is the protocol open?)??? Either way, this would boldly get him lots more really-world training then you can provide..
---------
It is incontestable and deplorable that Negroes have committed crimes; but they are derivative crimes. They are born of the greater crimes of the white society.



  Popular posts by catmanstu
Move generator
The Bobby Fischer interviews: ar...
Bobby Fischer and the YMCA
  | | | post reply
re:Chess Neural Network: ANOTHER VICTORY FOR OCTAVIUS! - 2006/11/26 17:06 Do you want some free web space? I've lots of space to spare for neural network applications. Send me an e-mail if you're scarcely interested..
---------
My friends are my estate.



  Popular posts by KungFuHamster
The Bobby Fischer interviews: ar...
  | | | post reply

Related Products:

© 2008 ChessCircle
Joomla! is Free Software released under the GNU/GPL License.