Implicit regularization in nonconvex statistical estimation: Gradient descent converges linearly for phase retrieval and matrix completion

C Ma, K Wang, Y Chi, Y Chen - International Conference on …, 2018‏ - proceedings.mlr.press
Recent years have seen a flurry of activities in designing provably efficient nonconvex
optimization procedures for solving statistical estimation problems. For various problems like …

The global optimization geometry of low-rank matrix optimization

Z Zhu, Q Li, G Tang, MB Wakin - IEEE Transactions on …, 2021‏ - ieeexplore.ieee.org
This paper considers general rank-constrained optimization problems that minimize a
general objective function over the set of rectangular matrices that have rank at most r. To …

A nonconvex approach for exact and efficient multichannel sparse blind deconvolution

Q Qu, X Li, Z Zhu - Advances in neural information …, 2019‏ - proceedings.neurips.cc
We study the multi-channel sparse blind deconvolution (MCS-BD) problem, whose task is to
simultaneously recover a kernel $\mathbf a $ and multiple sparse inputs $\{\mathbf x_i\} _ {i …

Manifold gradient descent solves multi-channel sparse blind deconvolution provably and efficiently

L Shi, Y Chi - IEEE Transactions on Information Theory, 2021‏ - ieeexplore.ieee.org
Multi-channel sparse blind deconvolution, or convolutional sparse coding, refers to the
problem of learning an unknown filter by observing its circulant convolutions with multiple …

Convex and nonconvex optimization are both minimax-optimal for noisy blind deconvolution under random designs

Y Chen, J Fan, B Wang, Y Yan - Journal of the American Statistical …, 2023‏ - Taylor & Francis
We investigate the effectiveness of convex relaxation and nonconvex optimization in solving
bilinear systems of equations under two different designs (ie, a sort of random Fourier …

Convolutional phase retrieval via gradient descent

Q Qu, Y Zhang, YC Eldar… - IEEE Transactions on …, 2019‏ - ieeexplore.ieee.org
We study the convolutional phase retrieval problem, of recovering an unknown signal x∈ C
n from m measurements consisting of the magnitude of its cyclic convolution with a given …

Recursive generalized extended least squares and RML algorithms for identification of bilinear systems with ARMA noise

Z Hafezi, MM Arefi - ISA transactions, 2019‏ - Elsevier
Bilinear systems are considered as a particular class of nonlinear systems including the
state variables which are typically used for online identification. By using a recursive …

Modified log-Sobolev inequalities, Beckner inequalities and moment estimates

R Adamczak, B Polaczyk, M Strzelecki - Journal of Functional Analysis, 2022‏ - Elsevier
We prove that in the context of general Markov semigroups Beckner inequalities with
constants separated from zero as p→ 1+ are equivalent to the modified log Sobolev …

Multi-Antenna ISAC Receiver with n-Tuple Blind Deconvolution

R Jacome, E Vargas, KV Mishra… - ICASSP 2024-2024 …, 2024‏ - ieeexplore.ieee.org
Recent developments in spectrum-sharing technologies include integrated sensing and
communications (ISAC) systems to save resources, cost, and power. In this paper, we …

Finding the sparsest vectors in a subspace: Theory, algorithms, and applications

Q Qu, Z Zhu, X Li, MC Tsakiris, J Wright… - arxiv preprint arxiv …, 2020‏ - arxiv.org
The problem of finding the sparsest vector (direction) in a low dimensional subspace can be
considered as a homogeneous variant of the sparse recovery problem, which finds …