Implicit regularization in nonconvex statistical estimation: gradient descent converges linearly for phase retrieval, matrix completion, and blind deconvolution (Q2189396): Difference between revisions

From MaRDI portal
Changed an Item
Import241208061232 (talk | contribs)
Normalize DOI.
 
(5 intermediate revisions by 5 users not shown)
Property / DOI
 
Property / DOI: 10.1007/s10208-019-09429-9 / rank
Normal rank
 
Property / MaRDI profile type
 
Property / MaRDI profile type: MaRDI publication profile / rank
 
Normal rank
Property / OpenAlex ID
 
Property / OpenAlex ID: W2768927505 / rank
 
Normal rank
Property / arXiv ID
 
Property / arXiv ID: 1711.10467 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Blind Deconvolution Using Convex Programming / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q3503433 / rank
 
Normal rank
Property / cites work
 
Property / cites work: A flexible convex relaxation for phase retrieval / rank
 
Normal rank
Property / cites work
 
Property / cites work: Non-Convex Phase Retrieval From STFT Measurements / rank
 
Normal rank
Property / cites work
 
Property / cites work: 10.1162/153244302760200704 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q2809807 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Fast rank-one alternating minimization algorithm for phase retrieval / rank
 
Normal rank
Property / cites work
 
Property / cites work: ROP: matrix recovery via rank-one projections / rank
 
Normal rank
Property / cites work
 
Property / cites work: Optimal rates of convergence for noisy sparse phase retrieval via thresholded Wirtinger flow / rank
 
Normal rank
Property / cites work
 
Property / cites work: A Probabilistic and RIPless Theory of Compressed Sensing / rank
 
Normal rank
Property / cites work
 
Property / cites work: The Power of Convex Relaxation: Near-Optimal Matrix Completion / rank
 
Normal rank
Property / cites work
 
Property / cites work: Phase Retrieval via Matrix Completion / rank
 
Normal rank
Property / cites work
 
Property / cites work: Solving quadratic equations via phaselift when there are about as many equations as unknowns / rank
 
Normal rank
Property / cites work
 
Property / cites work: Robust principal component analysis? / rank
 
Normal rank
Property / cites work
 
Property / cites work: Phase Retrieval via Wirtinger Flow: Theory and Algorithms / rank
 
Normal rank
Property / cites work
 
Property / cites work: Exact matrix completion via convex optimization / rank
 
Normal rank
Property / cites work
 
Property / cites work: Rank-Sparsity Incoherence for Matrix Decomposition / rank
 
Normal rank
Property / cites work
 
Property / cites work: Phase retrieval with one or two diffraction patterns by alternating projections with the null initialization / rank
 
Normal rank
Property / cites work
 
Property / cites work: Incoherence-Optimal Matrix Completion / rank
 
Normal rank
Property / cites work
 
Property / cites work: The Projected Power Method: An Efficient Algorithm for Joint Alignment from Pairwise Differences / rank
 
Normal rank
Property / cites work
 
Property / cites work: Solving Random Quadratic Systems of Equations Is Nearly as Easy as Solving Linear Systems / rank
 
Normal rank
Property / cites work
 
Property / cites work: Gradient descent with random initialization: fast global convergence for nonconvex phase retrieval / rank
 
Normal rank
Property / cites work
 
Property / cites work: Exact and Stable Covariance Estimation From Quadratic Sampling via Convex Programming / rank
 
Normal rank
Property / cites work
 
Property / cites work: Spectral method and regularized MLE are both optimal for top-\(K\) ranking / rank
 
Normal rank
Property / cites work
 
Property / cites work: Inference and uncertainty quantification for noisy matrix completion / rank
 
Normal rank
Property / cites work
 
Property / cites work: Nonconvex Optimization Meets Low-Rank Matrix Factorization: An Overview / rank
 
Normal rank
Property / cites work
 
Property / cites work: The Rotation of Eigenvectors by a Perturbation. III / rank
 
Normal rank
Property / cites work
 
Property / cites work: A note on \(\sin\Theta\) theorems for singular subspace variations / rank
 
Normal rank
Property / cites work
 
Property / cites work: On robust regression with high-dimensional predictors / rank
 
Normal rank
Property / cites work
 
Property / cites work: PhaseMax: Convex Phase Retrieval via Basis Pursuit / rank
 
Normal rank
Property / cites work
 
Property / cites work: Recovering Low-Rank Matrices From Few Coefficients in Any Basis / rank
 
Normal rank
Property / cites work
 
Property / cites work: An elementary proof of convex phase retrieval in the natural parameter space via the linear program PhaseMax / rank
 
Normal rank
Property / cites work
 
Property / cites work: Matrix Completion and Low-Rank SVD via Fast Alternating Least Squares / rank
 
Normal rank
Property / cites work
 
Property / cites work: Estimating the matrix \(p\)-norm / rank
 
Normal rank
Property / cites work
 
Property / cites work: A tail inequality for quadratic forms of subgaussian random vectors / rank
 
Normal rank
Property / cites work
 
Property / cites work: Blind Deconvolution by a Steepest Descent Algorithm on a Quotient Manifold / rank
 
Normal rank
Property / cites work
 
Property / cites work: Low-rank matrix completion using alternating minimization / rank
 
Normal rank
Property / cites work
 
Property / cites work: Debiasing the Lasso: optimal sample size for Gaussian designs / rank
 
Normal rank
Property / cites work
 
Property / cites work: Matrix Completion From a Few Entries / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q2896128 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Oracle inequalities in empirical risk minimization and sparse recovery problems. École d'Été de Probabilités de Saint-Flour XXXVIII-2008. / rank
 
Normal rank
Property / cites work
 
Property / cites work: Nuclear-norm penalization and optimal rates for noisy low-rank matrix completion / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q4695311 / rank
 
Normal rank
Property / cites work
 
Property / cites work: ADMiRA: Atomic Decomposition for Minimum Rank Approximation / rank
 
Normal rank
Property / cites work
 
Property / cites work: Blind Recovery of Sparse Signals From Subsampled Convolution / rank
 
Normal rank
Property / cites work
 
Property / cites work: Fast and Guaranteed Blind Multichannel Deconvolution Under a Bilinear System Model / rank
 
Normal rank
Property / cites work
 
Property / cites work: Fast, robust and non-convex subspace recovery / rank
 
Normal rank
Property / cites work
 
Property / cites work: Rapid, robust, and reliable blind deconvolution via nonconvex optimization / rank
 
Normal rank
Property / cites work
 
Property / cites work: Symmetry, Saddle Points, and Global Optimization Landscape of Nonconvex Matrix Factorization / rank
 
Normal rank
Property / cites work
 
Property / cites work: Self-calibration and biconvex compressive sensing / rank
 
Normal rank
Property / cites work
 
Property / cites work: Regularized gradient descent: a non-convex recipe for fast joint blind deconvolution and demixing / rank
 
Normal rank
Property / cites work
 
Property / cites work: The spectral norm of a nonnegative matrix / rank
 
Normal rank
Property / cites work
 
Property / cites work: Perturbation Bounds for the Polar Decomposition / rank
 
Normal rank
Property / cites work
 
Property / cites work: A Well-Tempered Landscape for Non-convex Robust Subspace Recovery / rank
 
Normal rank
Property / cites work
 
Property / cites work: The landscape of empirical risk for nonconvex losses / rank
 
Normal rank
Property / cites work
 
Property / cites work: Fundamental limits of weak recovery with applications to phase retrieval / rank
 
Normal rank
Property / cites work
 
Property / cites work: Restricted strong convexity and weighted matrix completion: Optimal bounds with noise / rank
 
Normal rank
Property / cites work
 
Property / cites work: Phase Retrieval Using Alternating Minimization / rank
 
Normal rank
Property / cites work
 
Property / cites work: A Simpler Approach to Matrix Completion / rank
 
Normal rank
Property / cites work
 
Property / cites work: Guaranteed Minimum-Rank Solutions of Linear Matrix Equations via Nuclear Norm Minimization / rank
 
Normal rank
Property / cites work
 
Property / cites work: Hanson-Wright inequality and sub-Gaussian concentration / rank
 
Normal rank
Property / cites work
 
Property / cites work: The local convexity of solving systems of quadratic equations / rank
 
Normal rank
Property / cites work
 
Property / cites work: Perturbation bounds for matrix square roots and Pythagorean sums / rank
 
Normal rank
Property / cites work
 
Property / cites work: Compressive Phase Retrieval via Generalized Approximate Message Passing / rank
 
Normal rank
Property / cites work
 
Property / cites work: Concentration and Moment Inequalities for Polynomials of Independent Random Variables / rank
 
Normal rank
Property / cites work
 
Property / cites work: GESPAR: Efficient Phase Retrieval of Sparse Signals / rank
 
Normal rank
Property / cites work
 
Property / cites work: Structured Signal Recovery From Quadratic Measurements: Breaking Sample Complexity Barriers via Nonconvex Optimization / rank
 
Normal rank
Property / cites work
 
Property / cites work: Theoretical Insights Into the Optimization Landscape of Over-Parameterized Shallow Neural Networks / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q4614113 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Complete Dictionary Recovery Over the Sphere I: Overview and the Geometric Picture / rank
 
Normal rank
Property / cites work
 
Property / cites work: Guaranteed Matrix Completion via Non-Convex Factorization / rank
 
Normal rank
Property / cites work
 
Property / cites work: The likelihood ratio test in high-dimensional logistic regression is asymptotically a rescaled Chi-square / rank
 
Normal rank
Property / cites work
 
Property / cites work: Phase retrieval via randomized Kaczmarz: theoretical guarantees / rank
 
Normal rank
Property / cites work
 
Property / cites work: Low rank matrix completion by alternating steepest descent methods / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q5388761 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Orthogonal Procrustes rotation for two or more matrices / rank
 
Normal rank
Property / cites work
 
Property / cites work: Convex Recovery of a Structured Signal from Independent Random Linear Measurements / rank
 
Normal rank
Property / cites work
 
Property / cites work: An Introduction to Matrix Concentration Inequalities / rank
 
Normal rank
Property / cites work
 
Property / cites work: Solving Systems of Random Quadratic Equations via Truncated Amplitude Flow / rank
 
Normal rank
Property / cites work
 
Property / cites work: Sparse Phase Retrieval via Truncated Amplitude Flow / rank
 
Normal rank
Property / cites work
 
Property / cites work: Perturbation bounds in connection with singular value decomposition / rank
 
Normal rank
Property / cites work
 
Property / cites work: Solving systems of phaseless equations via Kaczmarz methods: a proof of concept study / rank
 
Normal rank
Property / cites work
 
Property / cites work: Guarantees of Riemannian Optimization for Low Rank Matrix Recovery / rank
 
Normal rank
Property / cites work
 
Property / cites work: A useful variant of the Davis–Kahan theorem for statisticians / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q4637073 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Near-Optimal Bounds for Phase Synchronization / rank
 
Normal rank
Property / Wikidata QID
 
Property / Wikidata QID: Q127406705 / rank
 
Normal rank
Property / DOI
 
Property / DOI: 10.1007/S10208-019-09429-9 / rank
 
Normal rank

Latest revision as of 10:06, 17 December 2024

scientific article
Language Label Description Also known as
English
Implicit regularization in nonconvex statistical estimation: gradient descent converges linearly for phase retrieval, matrix completion, and blind deconvolution
scientific article

    Statements

    Implicit regularization in nonconvex statistical estimation: gradient descent converges linearly for phase retrieval, matrix completion, and blind deconvolution (English)
    0 references
    0 references
    0 references
    0 references
    0 references
    15 June 2020
    0 references
    nonconvex optimization
    0 references
    gradient descent
    0 references
    leave-one-out analysis
    0 references
    phase retrieval
    0 references
    matrix completion
    0 references
    blind deconvolution
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references

    Identifiers