Poster
Spectrally-normalized margin bounds for neural networks
Peter Bartlett · Dylan J Foster · Matus Telgarsky

Wed Dec 6th 06:30 -- 10:30 PM @ Pacific Ballroom #206 #None
in Posters Wed »

This paper presents a margin-based multiclass generalization bound for neural networks that scales with their margin-normalized "spectral complexity": their Lipschitz constant, meaning the product of the spectral norms of the weight matrices, times a certain correction factor. This bound is empirically investigated for a standard AlexNet network trained with SGD on the MNIST and CIFAR10 datasets, with both original and random labels; the bound, the Lipschitz constants, and the excess risks are all in direct correlation, suggesting both that SGD selects predictors whose complexity scales with the difficulty of the learning task, and secondly that the presented bound is sensitive to this complexity.

Keywords: Deep Learning · Learning Theory

Author Information

Peter Bartlett (UC Berkeley)
Dylan J Foster (Cornell University)
Matus Telgarsky (UIUC)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors