PAC-Bayesian Margin Bounds for Convolutional Neural Networks
arXiv:1801.00171
Abstract
Recently the generalization error of deep neural networks has been analyzed through the PAC-Bayesian framework, for the case of fully connected layers. We adapt this approach to the convolutional setting.
arXiv admin note: text overlap with arXiv:1707.09564 by other authors
References in corpus (5)
- A PAC-Bayesian Approach to Spectrally-Normalized Margin Bounds for Neural Networks
- Nearly-tight VC-dimension and pseudodimension bounds for piecewise linear neural networks
- Geometry of Optimization and Implicit Regularization in Deep Learning
- Data-Dependent Stability of Stochastic Gradient Descent
- FeTa: A DCA Pruning Algorithm with Generalization Error Guarantees
Cited by in corpus (5)
- Fantastic Generalization Measures and Where to Find Them
- PAC-Bayes with Backprop
- The intriguing role of module criticality in the generalization of deep networks
- How Many Samples are Needed to Estimate a Convolutional or Recurrent Neural Network?
- Sample Complexity Bounds for Recurrent Neural Networks with Application to Combinatorial Graph Problems