| Автор | R Meir |
| Автор | J F Fontanari |
| Дата выпуска | 1992-03-07 |
| dc.description | The authors study learning and generalization in single-layer feedforward networks, whose weights are constrained to take on a discrete set of values. Their analytic results are obtained within the replica approach, which is verified through Monte Carlo simulations. It is shown that, depending on the architecture of the network and on the source of the training examples, three qualitatively different behaviours emerge. This distinction, which is manifested through the dependence of the training and generalization errors on the size of the training set suggests a possible way to determine the suitability of the architecture to the learning task. They conjecture that this distinction is relevant to the more interesting case of multi-layered networks. |
| Формат | application.pdf |
| Издатель | Institute of Physics Publishing |
| Название | Learning from examples in weight-constrained neural networks |
| Тип | paper |
| DOI | 10.1088/0305-4470/25/5/021 |
| Print ISSN | 0305-4470 |
| Журнал | Journal of Physics A: Mathematical and General |
| Том | 25 |
| Первая страница | 1149 |
| Последняя страница | 1168 |
| Аффилиация | R Meir; Bellcore, Morristown, NJ, USA |
| Аффилиация | J F Fontanari; Bellcore, Morristown, NJ, USA |
| Выпуск | 5 |
