English

What training reveals about neural network complexity

Machine Learning 2021-11-01 v2 Machine Learning

Abstract

This work explores the Benevolent Training Hypothesis (BTH) which argues that the complexity of the function a deep neural network (NN) is learning can be deduced by its training dynamics. Our analysis provides evidence for BTH by relating the NN's Lipschitz constant at different regions of the input space with the behavior of the stochastic training procedure. We first observe that the Lipschitz constant close to the training data affects various aspects of the parameter trajectory, with more complex networks having a longer trajectory, bigger variance, and often veering further from their initialization. We then show that NNs whose 1st layer bias is trained more steadily (i.e., slowly and with little variation) have bounded complexity even in regions of the input space that are far from any training point. Finally, we find that steady training with Dropout implies a training- and data-dependent generalization bound that grows poly-logarithmically with the number of parameters. Overall, our results support the intuition that good training behavior can be a useful bias towards good generalization.

Keywords

Cite

@article{arxiv.2106.04186,
  title  = {What training reveals about neural network complexity},
  author = {Andreas Loukas and Marinos Poiitis and Stefanie Jegelka},
  journal= {arXiv preprint arXiv:2106.04186},
  year   = {2021}
}

Comments

Published at NeurIPS 2021

R2 v1 2026-06-24T02:56:55.738Z