Pith. sign in

REVIEW 2 cited by

High-dimensional limit theorems for SGD: Effective dynamics and critical scaling

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 2206.04030 v4 pith:GXAACNJW submitted 2022-06-08 stat.ML cs.LGmath.PRmath.STstat.TH

classification stat.MLcs.LGmath.PRmath.STstat.TH
keywords dynamicseffectivelimitlimitsscalingstep-sizeapproachballistic
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

We study the scaling limits of stochastic gradient descent (SGD) with constant step-size in the high-dimensional regime. We prove limit theorems for the trajectories of summary statistics (i.e., finite-dimensional functions) of SGD as the dimension goes to infinity. Our approach allows one to choose the summary statistics that are tracked, the initialization, and the step-size. It yields both ballistic (ODE) and diffusive (SDE) limits, with the limit depending dramatically on the former choices. We show a critical scaling regime for the step-size, below which the effective ballistic dynamics matches gradient flow for the population loss, but at which, a new correction term appears which changes the phase diagram. About the fixed points of this effective dynamics, the corresponding diffusive limits can be quite complex and even degenerate. We demonstrate our approach on popular examples including estimation for spiked matrix and tensor models and classification via two-layer networks for binary and XOR-type Gaussian mixture models. These examples exhibit surprising phenomena including multimodal timescales to convergence as well as convergence to sub-optimal solutions with probability bounded away from zero from random (e.g., Gaussian) initializations. At the same time, we demonstrate the benefit of overparametrization by showing that the latter probability goes to zero as the second layer width grows.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

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

  1. Benign Overfitting Does Not Occur in Diffusion Models

    stat.ML 2026-07 conditional novelty 7.0 of 10

    Benign overfitting and double descent do not occur in diffusion models: population and empirical score-matching losses cannot both be small without exponentially many samples.

  2. Specialization of softmax attention heads: insights from the high-dimensional single-location model

    cs.LG 2026-03 conditional novelty 7.0 of 10

    In a high-dimensional toy task, multi-head softmax attention first aligns all heads with the mean signal, then sequentially specializes to latent directions; the paper introduces Bayes-softmax, which attains the Bayes...

Pith tools