Weight Space Structure and Internal Representations: a Direct Approach to Learning and Generalization in Multilayer Neural Network

Details Weight Space Structure and Internal Representations: a Direct Approach to Learning and Generalization in Multilayer Neural Network Weight Space Structure and Internal Representations: a Direct Approach to Learning and Generalization in Multilayer Neural Network

1995-01-18

arxiv.org/abs/cond-mat/9501082v1

Physics

Condensed Matter

12 pages, 1 compressed ps figure (uufile), RevTeX file

Scientific paper

10.1103/PhysRevLett.75.2432

We analytically derive the geometrical structure of the weight space in multilayer neural networks (MLN), in terms of the volumes of couplings associated to the internal representations of the training set. Focusing on the parity and committee machines, we deduce their learning and generalization capabilities both reinterpreting some known properties and finding new exact results. The relationship between our approach and information theory as well as the Mitchison--Durbin calculation is established. Our results are exact in the limit of a large number of hidden units, showing that MLN are a class of exactly solvable models with a simple interpretation of replica symmetry breaking.

Affiliated with

Also associated with

No associations

Rating

Weight Space Structure and Internal Representations: a Direct Approach to Learning and Generalization in Multilayer Neural Network 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 Weight Space Structure and Internal Representations: a Direct Approach to Learning and Generalization in Multilayer Neural Network, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Weight Space Structure and Internal Representations: a Direct Approach to Learning and Generalization in Multilayer Neural Network will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-2175