Concepedia

Norm-Based Capacity Control in Neural Networks

Behnam Neyshabur, Ryota Tomioka, Nathan Srebro

Conference on Learning Theory · 2015 · 272 citations · 12 references

Concepts

Abstract

We investigate the capacity, convexity and characterization of a general family of norm-constrained feed-forward networks.

References

12

Rectified Linear Units Improve Restricted Boltzmann Machines

Vinod Nair, Geoffrey E. Hinton · International Conference on Machine Learning · 2010 · 13.2K citations

, , +16

The sample complexity of pattern classification with neural networks: the size of the weights is more important than the size of the network

Peter L. Bartlett · IEEE Transactions on Information Theory · 1998 · 1.2K citations · Full text

, , +12

Kernel Methods for Deep Learning

Youngmin Cho, Lawrence K. Saul · eScholarship (California Digital Library) · 2009 · 495 citations · Full text

, , +12

Empirical Margin Distributions and Bounding the Generalization Error of Combined Classifiers

Vladimir Koltchinskii, Dmitry Panchenko · The Annals of Statistics · 2002 · 475 citations · Full text

, , +17

On rectified linear units for speech processing

Matthew D. Zeiler, M. Ranzato, Rajat Monga et al. · 2013 · 420 citations

, , +14