Abstract | ||
---|---|---|
We study how certain smoothness constraints, for example, piecewise continuity, can be generalized from a discrete set of analog-valued data, by modifying the error backpropagation, learning algorithm. Numerical simulations demonstrate that by imposing two heuristic objectives - (1) reducing the number of hidden units, and (2) minimizing the magnitudes of the weights in the network - during the learning process, one obtains a network with a response function that smoothly interpolates between the training data. |
Year | DOI | Venue |
---|---|---|
1990 | 10.1162/neco.1990.2.2.188 | Neural Computation |
Field | DocType | Volume |
Training set,Mathematical optimization,Heuristic,Generalization,Artificial intelligence,Generalization error,Backpropagation,Smoothness,Machine learning,Piecewise,Mathematics | Journal | 2 |
Issue | ISSN | Citations |
2 | 0899-7667 | 17 |
PageRank | References | Authors |
45.64 | 2 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Chuanyi Ji | 1 | 812 | 124.04 |
Robert R. Snapp | 2 | 56 | 52.96 |
Demetri Psaltis | 3 | 431 | 209.24 |