Re: [computer-go] Pattern matching - example play

[Richard Brown <rbrown@xxxxxxxx>]

Thank you for the link!

Imran Ghory wrote:

> BTW. If you would like to know more about the application of TD algorithms
> and neural networks to games see my paper at,
>
> http://www.cs.bris.ac.uk/Publications/pub_info.jsp?id=2000100
>
> Particularly section 9 might interest you.
I haven't got to section 9 yet, but I am very impressed with this work.

Aside from the explanations themselves being very clear, and
convincing, may I assume also that the paper has been _peer_reviewed_?

Peer review, for those unclear on the concept, is where others, typically
those who share your interests, look at what you've done, all the while
saying, "I simply don't believe and only if you prove it to me then I will
believe".

The bastards!

The author of a million-dollar go program, or even one "so good that it's
worth a small fortune", can probably afford the luxury of disdaining peer
review.

[And there's nothing egotistical about that, if you've got the goods.
More power to you, I say.  The proof of the pudding is in the eating.]

At the same time, I suspect that _most_ of the readership on this list,
just as I am, are sufficiently-doubting-enough Thomases to sincerely
appreciate the type of results that Imran, and other peer-reviewed authors,
have obtained.  It's nice to see claims backed up by such solid research.

I do think that pattern-recognition will be a major component of strong
go programs of the future.  Of whose authorship, it remains to be seen.

--
Richard L. Brown             Office of Information Services
Senior Unix Sysadmin         University of Wisconsin System
                            780 Regent St., Rm. 246
rbrown@xxxxxxxxxxxxxxxxx              Madison, WI  53715

_______________________________________________
computer-go mailing list
computer-go@xxxxxxxxxxxxxxxxx
http://www.computer-go.org/mailman/listinfo/computer-go/