>> Suppose I develop what I think is an improved feature, for example
>> a better influence function or some other.  I'd like to hear
>> people's thoughts on how to best & most quickly determine if it is
>> in fact an improvement. ...

I've an almost-finished article on this subject, and the advantages of
using Very Large Test Suites over simply playing games. I've been
working on it for 6 months now, though, so I'll post a URL if/when I
ever finish it to my satisfaction.

On 2011-02-20 06:58, Petr Baudis wrote:
> Another factor is time per game. Some people advise for very fast 
> games and/or limitation by playout. I do not think either is good
> idea. Your idea is no good if the program beats gnugo in 5% more
> cases, but takes 20% longer per move. ...

That depends on your goal. If it only takes 20% longer to improve 5%
against your reference opponent, and all your other ways (such as dumb
scaling) require doubling the time to get the same improvement, then
your idea can be thought of as Very Good.

Darren

-- 
Darren Cook, Software Researcher/Developer

http://dcook.org/work/ (About me and my work)
http://dcook.org/blogs.html (My blogs and articles)
_______________________________________________
Computer-go mailing list
[email protected]
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go

Reply via email to