Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

I have heard that only a single layer is needed for an ANN to fit any possible function (input to output). Is this true and where is this investigated/state/found? Then what is the advantage of having multiple layers?

share|improve this question
add comment

1 Answer

up vote 1 down vote accepted

Search for the subject: How many hidden layers should I use?

http://www.faqs.org/faqs/ai-faq/neural-nets/part3/

share|improve this answer
    
the material on the site is not written to modern standard IMHO. I would hope that someone who is aquainted with the subject would give a nice answer here for others to then use; as an improvement. Or supply the names of some good papers that they believe is a first source for this question. –  Vass May 23 '11 at 14:21
1  
@Vass - you may want to ask this over at metaoptimize.com/qa if you are still looking for a more rigorous answer. –  lewellen Jun 17 '11 at 21:32
    
amazing link!!! Thank you. How come I never heard of this site?? I love it, because it is theoretical and practical. –  Vass Jun 25 '11 at 13:44
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.