Pith. sign in

REVIEW 1 cited by

WNGrad: Learn the Learning Rate in Gradient Descent

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 1803.02865 v2 pith:PZLU33YB submitted 2018-03-07 stat.ML cs.AIcs.LGcs.NAmath.NAmath.OC

classification stat.MLcs.AIcs.LGcs.NAmath.NAmath.OC
keywords learningratebatchlossgradientfunctionnormalizationstochastic
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
abstract

Adjusting the learning rate schedule in stochastic gradient methods is an important unresolved problem which requires tuning in practice. If certain parameters of the loss function such as smoothness or strong convexity constants are known, theoretical learning rate schedules can be applied. However, in practice, such parameters are not known, and the loss function of interest is not convex in any case. The recently proposed batch normalization reparametrization is widely adopted in most neural network architectures today because, among other advantages, it is robust to the choice of Lipschitz constant of the gradient in loss function, allowing one to set a large learning rate without worry. Inspired by batch normalization, we propose a general nonlinear update rule for the learning rate in batch and stochastic gradient descent so that the learning rate can be initialized at a high value, and is subsequently decreased according to gradient observations along the way. The proposed method is shown to achieve robustness to the relationship between the learning rate and the Lipschitz constant, and near-optimal convergence rates in both the batch and stochastic settings ($O(1/T)$ for smooth loss in the batch setting, and $O(1/\sqrt{T})$ for convex loss in the stochastic setting). We also show through numerical evidence that such robustness of the proposed method extends to highly nonconvex and possibly non-smooth loss function in deep learning problems.Our analysis establishes some first theoretical understanding into the observed robustness for batch normalization and weight normalization.

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. prunAdag: an adaptive pruning-aware gradient method

    math.OC 2025-02 conditional novelty 6.0 of 10

    prunAdag separates parameters into optimisable and decreasable sets, updates them with Adagrad-like rules, and provably drives the average gradient norm to zero at rate O(log(k)/sqrt(k+1)).

Pith tools