摘要: Supervised online learning with an ensemble of students randomized by the choice initial conditions is analyzed. For case perceptron rule, asymptotically same improvement in generalization error compared to performance a single student found as Gibbs learning. more optimized rules, however, using yields no improvement. This explained showing that for any rule f transform exists, such has behavior students.
harvard.edu
arxiv.org
arxiv-vanity.com参考文章(8)
T. L. H Watkin, Optimal Learning with a Neural Network EPL. ,vol. 21, pp. 871- 876 ,(1993) , 10.1209/0295-5075/21/8/013
Michael Biehl, Peter Riegler, Christian Wöhler, Transient dynamics of on-line learning in two-layered neural networks Journal of Physics A. ,vol. 29, pp. 4769- 4780 ,(1996) , 10.1088/0305-4470/29/16/005
O Kinouchi, N Caticha, Optimal generalization in perceptions Journal of Physics A. ,vol. 25, pp. 6243- 6250 ,(1992) , 10.1088/0305-4470/25/23/020
Roberta Simonetti, Nestor Caticha, On-line learning in parity machines Journal of Physics A. ,vol. 29, pp. 4859- 4867 ,(1996) , 10.1088/0305-4470/29/16/012
G. Reents, R. Urbanczik, SELF-AVERAGING AND ON-LINE LEARNING Physical Review Letters. ,vol. 80, pp. 5445- 5448 ,(1998) , 10.1103/PHYSREVLETT.80.5445
Manfred Opper, On-line versus Off-line Learning from Random Examples: General Results Physical Review Letters. ,vol. 77, pp. 4671- 4674 ,(1996) , 10.1103/PHYSREVLETT.77.4671
Manfred Opper, David Haussler, Generalization performance of Bayes optimal classification algorithm for learning a perceptron. Physical Review Letters. ,vol. 66, pp. 2677- 2680 ,(1991) , 10.1103/PHYSREVLETT.66.2677
Osame Kinouchi, Nestor Caticha, Learning algorithm that gives the Bayes generalization limit for perceptrons. Physical Review E. ,vol. 54, pp. 54- 57 ,(1996) , 10.1103/PHYSREVE.54.R54