Abstract | ||
---|---|---|
In this paper we address the important problem of optimizing regularization parameters in neural network modeling. The suggested optimization scheme is an extended version of the recently presented algorithm [25]. The idea is to minimize an empirical estimate - like the cross-validation estimate - of the generalization error with respect to regularization parameters. This is done by employing a simple iterative gradient descent scheme using virtually no additional programming overhead compared to standard training. Experiments with feed-forward neural network models for time series prediction and classification tasks showed the viability and robustness of the algorithm. Moreover, we provided some simple theoretical examples in order to illustrate the potential and limitations of the proposed regularization framework. |
Year | Venue | Keywords |
---|---|---|
2012 | Neural Networks: Tricks of the Trade, this book is an outgrowth of a 1996 NIPS workshop | Adaptive Regularization,Neural Network Modeling |
DocType | Volume | ISSN |
Conference | 1524 | 0302-9743 |
ISBN | Citations | PageRank |
3-540-65311-2 | 0 | 0.34 |
References | Authors | |
0 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jan Larsen | 1 | 98 | 18.05 |
Claus Svarer | 2 | 115 | 39.44 |
Lars Nonboe Andersen | 3 | 11 | 1.35 |
Lars Kai Hansen | 4 | 2776 | 341.03 |