I. Parberry, "Circuit complexity and feedforward neural networks," in Mathematical Perspectives on Neural Networks, P. Smolensky, M. Mozer, and D. Rumelhart, Eds. Hillsdale, NJ: Lawrence Erlbaum, 1996, pp. 85--111. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
Generalization Properties of Modular Networks: Implementing .. - Leonardo Franco And (2001) (Correct)
....Universidad Nacional de C ordoba, 5000) C ordoba, Argentina (e mail: cannas famaf.unc.edu.ar) previous layer and T i is the activation threshold of the neuron i . Di erent complexity measures as order of predicate, entropy decreasing criteria, size of the network, etc. see [2] 5] [13]) together with the fact that every change of a single input bit produces a change in the output make the parity function a hard problem among boolean functions. The parity function is thus one of the most used functions for testing learning algorithms, because of its simple de nition but great ....
.... From the vast literature where the parity function is analyzed and compared to another functions we could cite [5] 9] 10] 12] 16] 18] The question about what is the minimal size network needed to compute parity has been addressed from the point of view of Circuit Complexity, see [13]. Impagliazzo et al. 11] have found that the N bit parity function with a single hidden layer needs at least N 1 2 hidden neurons, while the best known construction has O(N) neurons. Recently in [8] it was demonstrated up to N=4, that the minimum size of the hidden layer required to solve ....
I. Parberry, \Circuit Complexity and Feedforward Neural Networks", in Mathematical Perspectives on Neural Networks, P. Smolensky, M. Mozer, D. Rumelhart, Eds., Lawrence Erlbaum Associates, pp. 85-111, 1996.
Generalization Properties of Modular Networks: Implementing.. - Franco, Cannas (2001) (Correct)
No context found.
I. Parberry, "Circuit complexity and feedforward neural networks," in Mathematical Perspectives on Neural Networks, P. Smolensky, M. Mozer, and D. Rumelhart, Eds. Hillsdale, NJ: Lawrence Erlbaum, 1996, pp. 85--111.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC