Randomized numerical linear algebra: Foundations and algorithms
PG Martinsson, JA Tropp - Acta Numerica, 2020 - cambridge.org
This survey describes probabilistic algorithms for linear algebraic computations, such as
factorizing matrices and solving linear systems. It focuses on techniques that have a proven …
factorizing matrices and solving linear systems. It focuses on techniques that have a proven …
An introduction to matrix concentration inequalities
JA Tropp - Foundations and Trends® in Machine Learning, 2015 - nowpublishers.com
Random matrices now play a role in many areas of theoretical, applied, and computational
mathematics. Therefore, it is desirable to have tools for studying random matrices that are …
mathematics. Therefore, it is desirable to have tools for studying random matrices that are …
Living on the edge: Phase transitions in convex programs with random data
Recent research indicates that many convex optimization problems with random constraints
exhibit a phase transition as the number of constraints increases. For example, this …
exhibit a phase transition as the number of constraints increases. For example, this …
Sparse signal processing concepts for efficient 5G system design
As it becomes increasingly apparent that 4G will not be able to meet the emerging demands
of future mobile communication systems, the question what could make up a 5G system …
of future mobile communication systems, the question what could make up a 5G system …
Universality laws for randomized dimension reduction, with applications
Dimension reduction is the process of embedding high-dimensional data into a lower
dimensional space to facilitate its analysis. In the Euclidean setting, one fundamental …
dimensional space to facilitate its analysis. In the Euclidean setting, one fundamental …
The squared-error of generalized lasso: A precise analysis
S Oymak, C Thrampoulidis… - 2013 51st Annual Allerton …, 2013 - ieeexplore.ieee.org
We consider the problem of estimating an unknown but structured signal x 0 from its noisy
linear observations y= Ax 0+ z∈ ℝ m. To the structure of x 0 is associated a structure …
linear observations y= Ax 0+ z∈ ℝ m. To the structure of x 0 is associated a structure …
A new perspective on least squares under convex constraint
S Chatterjee - 2014 - projecteuclid.org
Consider the problem of estimating the mean of a Gaussian random vector when the mean
vector is assumed to be in a given convex set. The most natural solution is to take the …
vector is assumed to be in a given convex set. The most natural solution is to take the …
Beyond low rank+ sparse: Multiscale low rank matrix decomposition
We present a natural generalization of the recent low rank+ sparse matrix decomposition
and consider the decomposition of matrices into components of multiple scales. Such …
and consider the decomposition of matrices into components of multiple scales. Such …
Regularized gradient descent: a non-convex recipe for fast joint blind deconvolution and demixing
S Ling, T Strohmer - Information and Inference: A Journal of the …, 2019 - academic.oup.com
We study the question of extracting a sequence of functions from observing only the sum of
their convolutions, ie from. While convex optimization techniques are able to solve this joint …
their convolutions, ie from. While convex optimization techniques are able to solve this joint …
Blind demixing and deconvolution at near-optimal rate
We consider simultaneous blind deconvolution of r source signals from their noisy
superposition, a problem also referred to blind demixing and deconvolution. This signal …
superposition, a problem also referred to blind demixing and deconvolution. This signal …