A multifractal phase-space analysis of perceptrons with biased patterns

Details A multifractal phase-space analysis of perceptrons with biased patterns A multifractal phase-space analysis of perceptrons with biased patterns

1997-10-15

arxiv.org/abs/cond-mat/9710154v2

J. Phys. A: Math. Gen 31 (1998) 2509-2517

Physics

Condensed Matter

Disordered Systems and Neural Networks

7 pages, Latex, 2 eps-files, submitted to J. Phys. A, new title, enlarged description of the simulation algorithm, typos corre

Scientific paper

10.1088/0305-4470/31/11/004

We calculate the multifractal spectrum of the partition of the coupling space of a perceptron induced by random input-output pairs with non-zero mean. From the results we infer the influence of the input and output bias respectively on both the storage and generalization properties of the network. It turns out that the value of the input bias is irrelevant as long as it is different from zero. The generalization problem with output bias is new and shows an interesting two-level scenario. To compare our analytical results with simulations we introduce a simple and efficient algorithm to implement Gibbs learning.

Affiliated with

Also associated with

No associations

Rating

A multifractal phase-space analysis of perceptrons with biased patterns does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with A multifractal phase-space analysis of perceptrons with biased patterns, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and A multifractal phase-space analysis of perceptrons with biased patterns will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-166388