Learning in feed-forward neural networks by improving the performance
Statistical mechanics is used to derive a new learning rule for a feed-forward neural network with one hidden layer. Generalization to multilayer neural networks is straightforward, and proceeds in the same way as backpropagation.
Year of publication: |
1992
|
---|---|
Authors: | Gordon, Mirta B. ; Pereto, Pierre ; Rodriguez-Girones, Miguel |
Published in: |
Physica A: Statistical Mechanics and its Applications. - Elsevier, ISSN 0378-4371. - Vol. 185.1992, 1, p. 402-410
|
Publisher: |
Elsevier |
Saved in:
Saved in favorites
Similar items by person
-
Seller's dilemma due to social interactions between customers
Gordon, Mirta B., (2005)
-
A model for scaling in firms’ size and growth rate distribution
Metzig, Cornelia, (2014)
-
Collective states in social systems with interacting learning agents
Semeshenko, Viktoriya, (2008)
- More ...