[an error occurred while processing this directive] [an error occurred while processing this directive][an error occurred while processing this directive] [an error occurred while processing this directive] [an error occurred while processing this directive] [an error occurred while processing this directive] (none) [an error occurred while processing this directive] [an error occurred while processing this directive] [an error occurred while processing this directive] [an error occurred while processing this directive] [an error occurred while processing this directive][an error occurred while processing this directive] [an error occurred while processing this directive][an error occurred while processing this directive] [an error occurred while processing this directive][an error occurred while processing this directive] [an error occurred while processing this directive] [an error occurred while processing this directive] [an error occurred while processing this directive] (none) [an error occurred while processing this directive] [an error occurred while processing this directive] [an error occurred while processing this directive][an error occurred while processing this directive]
 
[an error occurred while processing this directive] [an error occurred while processing this directive]
Skåne Sjælland Linux User Group - http://www.sslug.dk Home   Subscribe   Mail Archive   Forum   Calendar   Search
MhonArc Date: [Date Prev] [Date Index] [Date Next]   Thread: [Date Prev] [Thread Index] [Date Next]   MhonArc
 

Re: [CPROG] Finde globalt max/min



> Glatte funktioner? SA evaluere kun funktionen i punkter, og du kan
> vælge dine hop som du ønsker, så funktiontioner behøver ikke engang
> at være kontinuert!

Det er rigtigt, men hele idéen med naboklasser er at man skal bevæge sig op og ned ad bakker i
energilandskabet. Det går ikke rigtigt uden en vis glathed.
Hvis støjen er af samme størelsesorden som selve signalet, så duer SA ikke (den vil fumle i
blinde).
 
> > Hvis problemet er som du beskriver (en mapping f:R->R) så er SA
> > at skyde gåspurve med kanoner.
> 
> Hvis funktionen er kontinuert og differentiabel så er gradient
> descent metoder bedrer,

Nej, det kan man ikke sige generelt, det kommer an på hvordan energilandskabet ser ud. Nogle
problemer har masser af lokale minimaer som gradient metoder vil sidde fast i. Det kan tildels
løses ved at køre med et stort ensemble, men kun SA (genetiske algo. og lign. metoder) kan vandre
frit.

> hvis funktionen ikke er differentiabel,
> eller hvis det er meget dyrt at udregne afledede så kunne man
> bruge SA.

Alting er differentiabelt numerisk set ;)

Jeg synes bare umiddelbart at det må være overkill at bruge SA på en skalarfunktion.

__________________________________________________
Do You Yahoo!?
Yahoo! Auctions - buy the things you want at great prices
http://auctions.yahoo.com/


 
Home   Subscribe   Mail Archive   Index   Calendar   Search

 
 
Questions about the web-pages to <www_admin>. Last modified 2005-08-10, 20:08 CEST [an error occurred while processing this directive]
This page is maintained by [an error occurred while processing this directive]MHonArc [an error occurred while processing this directive] # [an error occurred while processing this directive] *