Pith. sign in

REVIEW 3 cited by

A Randomised Subspace Gauss-Newton Method for Nonlinear Least-Squares

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 2211.05727 v1 pith:FD4K7ZBQ submitted 2022-11-10 math.OC cs.LG

classification math.OCcs.LG
keywords least-squaresnonlinearr-sgngauss-newtonpresentedproblemsrandomisedregression
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

We propose a Randomised Subspace Gauss-Newton (R-SGN) algorithm for solving nonlinear least-squares optimization problems, that uses a sketched Jacobian of the residual in the variable domain and solves a reduced linear least-squares on each iteration. A sublinear global rate of convergence result is presented for a trust-region variant of R-SGN, with high probability, which matches deterministic counterpart results in the order of the accuracy tolerance. Promising preliminary numerical results are presented for R-SGN on logistic regression and on nonlinear regression problems from the CUTEst collection.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 3 Pith papers

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

  1. Model-Driven Subspaces for Large-Scale Optimization with Local Approximation Strategy

    math.OC 2025-09 reject novelty 6.0 of 10

    The paper proposes truncated, model-gradient-generated subspaces for large-scale optimization and gives conditional decrease and convergence theorems, but the stated guarantees are not fully proven.

  2. Monotone and nonmonotone linearized block coordinate descent methods for nonsmooth composite optimization problems

    math.OC 2025-06 conditional novelty 6.0 of 10

    Two linearized block coordinate descent algorithms for nonsmooth composite optimization converge in expectation to a stationary point at O(1/ε²) rate.

  3. A variable dimension sketching strategy for nonlinear least-squares

    math.OC 2025-06 conditional novelty 6.0 of 10

    A randomized subspace Levenberg-Marquardt method with adaptively chosen subspace size retains O(epsilon^-2) complexity and shows practical cost savings.

Pith tools