Pith. sign in

REVIEW 7 cited by

Better Theory for SGD in the Nonconvex World

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 2002.03329 v3 pith:MZSRYVL6 submitted 2020-02-09 math.OC cs.LGstat.ML

Better Theory for SGD in the Nonconvex World

classification math.OC cs.LGstat.ML
keywords nonconvexanalysisassumptionconvergenceconvexityfindinggradientmathcal
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Large-scale nonconvex optimization problems are ubiquitous in modern machine learning, and among practitioners interested in solving them, Stochastic Gradient Descent (SGD) reigns supreme. We revisit the analysis of SGD in the nonconvex setting and propose a new variant of the recently introduced expected smoothness assumption which governs the behaviour of the second moment of the stochastic gradient. We show that our assumption is both more general and more reasonable than assumptions made in all prior work. Moreover, our results yield the optimal $\mathcal{O}(\varepsilon^{-4})$ rate for finding a stationary point of nonconvex smooth functions, and recover the optimal $\mathcal{O}(\varepsilon^{-1})$ rate for finding a global solution if the Polyak-{\L}ojasiewicz condition is satisfied. We compare against convergence rates under convexity and prove a theorem on the convergence of SGD under Quadratic Functional Growth and convexity, which might be of independent interest. Moreover, we perform our analysis in a framework which allows for a detailed study of the effects of a wide array of sampling strategies and minibatch sizes for finite-sum optimization problems. We corroborate our theoretical results with experiments on real and synthetic data.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 7 Pith papers

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

  1. LOSCAR-SGD: Local SGD with Communication-Computation Overlap and Delay-Corrected Sparse Model Averaging

    cs.LG 2026-05 unverdicted novelty 7.0

    LOSCAR-SGD combines local updates, sparse model averaging, and communication-computation overlap with a delay-corrected merge rule, providing convergence rates for smooth non-convex objectives under worker heterogeneity.

  2. Ringmaster LMO: Asynchronous Linear Minimization Oracle Momentum Method

    cs.LG 2026-05 unverdicted novelty 7.0

    Ringmaster LMO extends delay-thresholding from ASGD to LMO-based momentum updates, providing convergence guarantees under (L0, L1)-smoothness and time-complexity bounds that recover optimal rates in the Euclidean case.

  3. Select-then-differentiate: Solving Bilevel Optimization with Manifold Lower-level Solution Sets

    math.OC 2026-05 unverdicted novelty 7.0

    Optimistic bilevel optimization with manifold lower-level minimizers is differentiable if the optimistic selection is unique, yielding a pseudoinverse hyper-gradient and a convergent HG-MS algorithm whose rate depends...

  4. A stochastic gradient algorithm for non-separable optimization with convergence guarantee

    math.OC 2026-06 unverdicted novelty 6.0

    Presents a stochastic gradient algorithm for non-separable optimization with local convergence guarantees under smoothness assumptions.

  5. Rennala MVR: Improved Time Complexity for Parallel Stochastic Optimization via Momentum-Based Variance Reduction

    math.OC 2026-05 unverdicted novelty 5.0

    Rennala MVR improves time complexity over Rennala SGD for smooth nonconvex stochastic optimization in heterogeneous parallel systems under a mean-squared smoothness assumption.

  6. Topology-Aware Block Coordinate Descent for Qubit Frequency Allocation of Superconducting Quantum Processors

    quant-ph 2026-01 reject novelty 5.0

    The Snake optimizer is identified with block coordinate descent, and a nearest-neighbor heuristic over a sequence-dependent TSP is proposed to order blocks and reduce calibration cost; the SD-TSP cost is not specified.

  7. Stochastic versus Deterministic in Stochastic Gradient Descent

    math.OC 2025-09 unverdicted novelty 5.0

    Treating stochastic and deterministic gradients separately in mini-batch SGD yields faster convergence and smaller error radius than uniform treatment, with further gains under strong convexity.