##
1-Bit matrix completion

in** Information and Inference: A Journal of the IMA**

September 2014; p ublished online July 2014 .

Journal Article. Subjects: Science and Mathematics; Mathematics; Applied Mathematics; Computer Science. 0 words.

In this paper, we develop a theory of matrix completion for the extreme case of noisy 1-bit observations. Instead of observing a subset of the real-valued entries of a matrix **null...**

##
CGIHT: conjugate gradient iterative hard thresholding for compressed sensing and matrix completion

in** Information and Inference: A Journal of the IMA**

December 2015; p ublished online November 2015 .

Journal Article. Subjects: Science and Mathematics; Mathematics; Applied Mathematics; Computer Science. 16576 words.

We introduce the conjugate gradient iterative hard thresholding (CGIHT) family of algorithms for the efficient solution of constrained underdetermined linear systems of equations arising in...

##
Compressed subspace matching on the continuum

in** Information and Inference: A Journal of the IMA**

June 2015; p ublished online April 2015 .

Journal Article. Subjects: Science and Mathematics; Mathematics; Applied Mathematics; Computer Science. 12813 words.

We consider the general problem of matching a subspace to a signal in [math] that has been observed in-directly (compressed) through a random projection. We are interested in the case where...

##
Deep Haar scattering networks

in** Information and Inference: A Journal of the IMA**

June 2016; p ublished online April 2016 .

Journal Article. Subjects: Science and Mathematics; Mathematics; Applied Mathematics; Computer Science. 12296 words.

An orthogonal Haar scattering transform is a deep network computed with a hierarchy of additions, subtractions and absolute values over pairs of coefficients. Unsupervised learning...

##
Detecting the large entries of a sparse covariance matrix in sub-quadratic time

in** Information and Inference: A Journal of the IMA**

September 2016; p ublished online March 2016 .

Journal Article. Subjects: Science and Mathematics; Mathematics; Applied Mathematics; Computer Science. 10832 words.

The covariance matrix of a [math]-dimensional random variable is a fundamental quantity in data analysis. Given [math] i.i.d. observations, it is typically estimated by the sample...

##
Deterministic Bayesian information fusion and the analysis of its performance

in** Information and Inference: A Journal of the IMA**

December 2014; p ublished online December 2014 .

Journal Article. Subjects: Science and Mathematics; Mathematics; Applied Mathematics; Computer Science. 9764 words.

This paper develops a mathematical and computational framework for analyzing the expected performance of Bayesian data fusion, or joint statistical inference, within a sensor network. We...

##
Disparity and optical flow partitioning using extended Potts priors

in** Information and Inference: A Journal of the IMA**

March 2015; p ublished online December 2014 .

Journal Article. Subjects: Science and Mathematics; Mathematics; Applied Mathematics; Computer Science. 8344 words.

This paper addresses the problems of disparity and optical flow partitioning based on the brightness invariance assumption. We investigate new variational approaches to these problems with...

##
Finite sample posterior concentration in high-dimensional regression

in** Information and Inference: A Journal of the IMA**

March 2015; p ublished online October 2014 .

Journal Article. Subjects: Science and Mathematics; Mathematics; Applied Mathematics; Computer Science. 54 words.

##
Finite sample posterior concentration in high-dimensional regression

in** Information and Inference: A Journal of the IMA**

June 2014; p ublished online June 2014 .

Journal Article. Subjects: Science and Mathematics; Mathematics; Applied Mathematics; Computer Science. 0 words.

We study the behavior of the posterior distribution in high-dimensional Bayesian Gaussian linear regression models having *p* ≫ *n*, where *p* is the number of predictors and *n* is the sample...

##
Graph connection Laplacian and random matrices with random blocks

in** Information and Inference: A Journal of the IMA**

March 2015; p ublished online March 2015 .

Journal Article. Subjects: Science and Mathematics; Mathematics; Applied Mathematics; Computer Science. 16604 words.

Graph connection Laplacian (GCL) is a modern data analysis technique that is starting to be applied for the analysis of high-dimensional and massive datasets. Motivated by this technique,...