Storage Capacity of the Tilinglike Learning Algorithm
arXiv:cond-mat/0008162 · doi:10.1063/1.1358157
Abstract
The storage capacity of an incremental learning algorithm for the parity machine, the Tilinglike Learning Algorithm, is analytically determined in the limit of a large number of hidden perceptrons. Different learning rules for the simple perceptron are investigated. The usual Gardner-Derrida one leads to a storage capacity close to the upper bound, which is independent of the learning algorithm considered.
Proceedings of the Conference Disordered and Complex Systems, King's College, London, July 2000. 6 pages, 1 figure, uses aipproc.sty