Abstract | ||
---|---|---|
Exhaustive exploration of an ensemble of networks is used to model learning and generalization in layered neural networks. A simple Boolean learning problem involving networks with binary weights is numerically solved to obtain the entropy Sm and the average generalization ability Gm as a function of the size m of the training set. Learning curves Gm vs m are shown to depend solely on the distribution of generalization abilities over the ensemble of networks. Such distribution is determined prior to learning, and provides a novel theoretical tool for the prediction of network performance on a specific task. |
Year | DOI | Venue |
---|---|---|
1990 | 10.1162/neco.1990.2.3.374 | Neural Computation |
Keywords | DocType | Volume |
novel theoretical tool,network performance,exhaustive learning,layered neural network,size m,generalization ability,exhaustive exploration,binary weight,entropy sm,average generalization ability gm,curves gm | Journal | 2 |
Issue | ISSN | Citations |
3 | 0899-7667 | 17 |
PageRank | References | Authors |
33.98 | 3 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
D. B. Schwartz | 1 | 17 | 34.32 |
V. K. Samalam | 2 | 30 | 35.41 |
Sara A. Solla | 3 | 1095 | 1249.18 |
J. S. Denker | 4 | 3245 | 2524.81 |