Gradient descent with random initialization: fast global convergence for nonconvex phase retrieval
Open accessThe DOI matches its Crossref record; 2 of the 2 references it cites were found in a registry or an index.
Abstract
This paper considers the problem of solving systems of quadratic equations, namely, recovering an object of interest x♮∈Rn\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varvec{x}^{\natural }\in {\mathbb {R}}^{n}$$\end{document} from m quadratic equations/samples yi=(ai⊤x♮)2,1≤i≤m\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$y_{i}=(\varvec{a}_{i}^{\top }\varvec{x}^{\natural })^{2}, 1\le i\le m$$\end{document}. This problem, also dubbed as phase retrieval, spans multiple domains including physical sciences and machine learning. We investigate the efficacy of gradient descent (or Wirtinger flow) designed for the nonconvex least squares problem. We prove that under Gaussian designs, gradient descent—when randomly initialized—yields an ϵ\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\epsilon $$\end{document}-accurate solution in O(logn+log(1/ϵ))\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$O\big (\log n+\log (1/\epsilon )\big )$$\end{document} iterations given nearly minimal samples, thus achieving near-optimal computational and sample complexities at once. This provides the first global convergence guarantee concerning vanilla gradient descent for phase retrieval, without the need of (i) carefully-designed initialization, (ii) sample splitting, or (iii) sophisticated saddle-point escaping schemes. All of these are achieved by exploiting the statistical models in analyzing optimization algorithms, via a leave-one-out approach that enables the decoupling of certain statistical dependency between the gradient descent iterates and the data.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Nonconvex Optimization Meets Low-Rank Matrix Factorization: An Overview
Chi, Y., Lu, Y. M., & Chen, Y. (2019) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2019.2937282 · Cited by 360
Self-citationTheoretical Insights Into the Optimization Landscape of Over-Parameterized Shallow Neural Networks
Soltanolkotabi, M., Javanmard, A., & Lee, J. D. (2019) · IEEE Transactions on Information Theory · doi:10.1109/tit.2018.2854560 · Cited by 356
Cited by
Neural holography with camera-in-the-loop training
Peng, Y., Choi, S., Padmanaban, N., & Wetzstein, G. (2020) · ACM Transactions on Graphics · doi:10.1145/3414685.3417802 · Cited by 514
Quantum Generative Adversarial Networks for learning and loading random distributions
Zoufal, C., Lucchi, A., & Woerner, S. (2019) · npj Quantum Information · doi:10.1038/s41534-019-0223-2 · Cited by 480
Nonconvex Optimization Meets Low-Rank Matrix Factorization: An Overview
Chi, Y., Lu, Y. M., & Chen, Y. (2019) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2019.2937282 · Cited by 360
“Later on, GD was shown to converge within a logarithmic number of iterations for phase retrieval, even with random initialization [74].”
BackgroundSelf-citationBlind Over-the-Air Computation and Data Fusion via Provable Wirtinger Flow
Dong, J., Shi, Y., & Ding, Z. (2020) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2020.2970338 · Cited by 65
Subspace Estimation from Unbalanced and Incomplete Data Matrices: 𝓁2, ∞ Statistical Guarantees
Cai, C., Li, G., Chi, Y., Poor, H. V., & Chen, Y. (2021) · The Annals of Statistics · doi:10.1214/20-aos1986 · Cited by 37
Self-citation
Metadata
- Authors
- Chen, Y., Chi, Y., Fan, J., & Ma, C.
- Venue
- Mathematical Programming
- Published
- 2019-02-04
- Volume
- 176
- Issue
- 1-2
- Pages
- 5-37
- ISSN
- 0025-5610 (ISSN-L)
- 1436-4646
- OpenAlex
- W2790687101
- Cited by
- 122
- References
- 2
- FWCI
- 11.11
- Type
- journal article
- Language
- en
- Open access
- green
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Gradient descent with random initialization: fast global convergence for nonconvex phase retrieval. https://app.cochat.ai/papers/10.1007/s10107-019-01363-6
- The paper
- Chen, Y., Chi, Y., Fan, J., & Ma, C. (2019). Gradient descent with random initialization: fast global convergence for nonconvex phase retrieval. Mathematical Programming. https://doi.org/10.1007/s10107-019-01363-6
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.