• Publications
  • Influence
Model-Based Compressive Sensing
TLDR
A model-based CS theory is introduced that parallels the conventional theory and provides concrete guidelines on how to create model- based recovery algorithms with provable performance guarantees and a new class of structured compressible signals along with a new sufficient condition for robust structured compressable signal recovery that is the natural counterpart to the restricted isometry property of conventional CS.
Bilinear Generalized Approximate Message Passing—Part I: Derivation
TLDR
This paper derives the Bilinear G-AMP (BiG-AMP) algorithm as an approximation of the sum-product belief propagation algorithm in the high-dimensional limit, where central-limit theorem arguments and Taylor-series approximations apply, and under the assumption of statistically independent matrix entries with known priors.
Sparse Signal Recovery Using Markov Random Fields
TLDR
This paper uses Markov Random Fields to represent sparse signals whose nonzero coefficients are clustered and uses a model-based recovery algorithm to stably recovers MRF-modeled signals using many fewer measurements and computations than the current state-of-the-art algorithms.
Practical Sketching Algorithms for Low-Rank Matrix Approximation
TLDR
A suite of algorithms for constructing low-rank approximations of an input matrix from a random linear image, or sketch, of the matrix that can preserve structural properties of the input matrix, such as positive-semidefiniteness, and they can produce approximation with a user-specified rank.
Convex Optimization for Big Data: Scalable, randomized, and parallel algorithms for big data analytics
TLDR
An overview of this emerging field, contemporary approximation techniques such as first-order methods and randomization for scalability, and the important role of parallel and distributed computation are described.
Adversarially Robust Optimization with Gaussian Processes
TLDR
It is shown that standard GP optimization algorithms do not exhibit the desired robustness properties, and a novel confidence-bound based algorithm StableOpt is provided for this purpose, which consistently succeeds in finding a stable maximizer where several baseline methods fail.
Time-Varying Gaussian Process Bandit Optimization
TLDR
This work introduces two natural extensions of the classical Gaussian process upper confidence bound (GP- UCB) algorithm, and finds the gradual forgetting of TV-GP-UCB to perform favorably compared to the sharp resetting of R-GP -UCB.
Compressive Sensing for Background Subtraction
TLDR
A method to directly recover background subtracted images using CS and its applications in some communication constrained multi-camera computer vision problems is described and its approach is suitable for image coding in communication constrained problems.
A compressive beamforming method
TLDR
This paper considers the direction-of-arrival (DOA) estimation problem with an array of sensors using CS and shows that by using random projections of the sensor data, along with a full waveform recording on one reference sensor, a sparse angle space scenario can be reconstructed, giving the number of sources and their DOA's.
WASP: Scalable Bayes via barycenters of subset posteriors
TLDR
The Wasserstein posterior (WASP) has an atomic form, facilitating straightforward estimation of posterior summaries of functionals of interest and theoretical justification in terms of posterior consistency and algorithm eciency is provided.
...
...