Pith. sign in

REVIEW 1 cited by

Towards Explaining the Regularization Effect of Initial Large Learning Rate in Training Neural Networks

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 1907.04595 v2 pith:DIMOVFKK submitted 2019-07-10 cs.LG stat.ML

classification cs.LGstat.ML
keywords learningratelargesmallinitialmodelbetterpatterns
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Stochastic gradient descent with a large initial learning rate is widely used for training modern neural net architectures. Although a small initial learning rate allows for faster training and better test performance initially, the large learning rate achieves better generalization soon after the learning rate is annealed. Towards explaining this phenomenon, we devise a setting in which we can prove that a two layer network trained with large initial learning rate and annealing provably generalizes better than the same network trained with a small learning rate from the start. The key insight in our analysis is that the order of learning different types of patterns is crucial: because the small learning rate model first memorizes easy-to-generalize, hard-to-fit patterns, it generalizes worse on hard-to-generalize, easier-to-fit patterns than its large learning rate counterpart. This concept translates to a larger-scale setting: we demonstrate that one can add a small patch to CIFAR-10 images that is immediately memorizable by a model with small initial learning rate, but ignored by the model with large learning rate until after annealing. Our experiments show that this causes the small learning rate model's accuracy on unmodified images to suffer, as it relies too much on the patch early on.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. SGD as Free Energy Minimization: A Thermodynamic View on Neural Network Training

    cs.LG 2025-05 reject novelty 5.0 of 10

    SGD is said to minimize free energy, but the temperature is constructed from the data, making the validation circular.

Pith tools