| Michael C. Mozer, Michael I. Jordan, Thomas Petsche - 1997 - 1128 lapas
...Proceedings of the CTP-PBSRI Joint Workshop on Theoretical Physics, J.-H. Oh et al eds, 105-130, 1995. For valid generalization, the size of the weights is more important than the size of the network Peter L. Bartlett Department of Systems Engineering Research School of Information Sciences and Engineering... | |
| Shai Ben-David - 1997 - 350 lapas
...207-217, (1993). 3. Peter Bartlett, "The Sample Complexity of Pattern Classification with Neural Networks: the Size of the Weights is More Important than the Size of the Network," Technical Report. Department of Systems Engineering, Australian National University, May 1996. 4. Bernhard... | |
| Bernhard Schölkopf, Christopher J. C. Burges, Alexander J. Smola - 1999 - 400 lapas
...44(2):525-536, 1998a. PL Bartlett. The sample complexity of pattern classification with neural networks: the size of the weights is more important than the size of the network. IEEE Transactions on Information Theory, 44(2):525-536, 1998b. PL Bartlett, P. Long, and RC Williamson.... | |
| Michael S. Kearns, Sara A. Solla, David A. Cohn - 1999 - 1122 lapas
...References [Bar98] PL Bartlett. The sample complexity of pattern classification with neural networks: the size of the weights is more important than the size of the network. IEEE Transactions on Information Theory, 44(2):525-536, 1998. [Bre97] L. Breiman. Prediction games... | |
| Alexander J. Smola - 2000 - 436 lapas
...54(2):317-331, 1997. PL Bartlett. The sample complexity of pattern classification with neural networks: the size of the weights is more important than the size of the network. IEEE Transactions on Information Theory, 44(2):525-536, 1998. PL Bartlett, SR Kulkarni, and SE Posner.... | |
| Sara A. Solla, Klaus-Robert Müller, Todd K. Leen - 2000 - 1124 lapas
...overfit. References [1] PL Bartlett. The sample complexity of pattern classification with neural networks: the size of the weights is more important than the size of the network. IEEE Transactions on Information Theory, 44(2):525-536, March 1998. [2] L. Breiman. Bagging predictors.... | |
| Todd K. Leen, Thomas G. Dietterich, Volker Tresp - 2001 - 1136 lapas
...training data, while BP is biased toward learning smoolher functions. References [ 1 ] Peter L. Bartlett. For valid generalization the size of the weights is more important than the size of the network. In Advances in Neural Information Processing Systems, volume 9, page 134. The MIT Press, 1997. [2]... | |
| Ralf Herbrich - 2001 - 402 lapas
...deGryter. Bartlett, PL (1998). The sample complexity of pattern classification with neural networks: The size of the weights is more important than the size of the network. IEEE Transactions on Information Theory 44(2), 525-536. Bartlett, PL and J. Shawe-Taylor(1999). Generalization... | |
| Ajith Abraham, Mario Köppen, Katrin Franke - 2003 - 1160 lapas
...weights, the smoother the building-blocks. And the reason behind Bartlett's well-known observation that "for valid generalization, the size of the weights is more important than the size of the network" is now crystal clear from the viewpoint of this geometrical interpretation. Various methods of preventing... | |
| |