activity
20122021
most citedProvable Online CP/PARAFAC Decomposition of a Structured Tensor via Dictionary Learning

8 citations · 27 across the 9 of their papers we have counts for

collaborators

16 papers

cs.LG20211 cited

Online Stochastic Gradient Descent Learns Linear Dynamical Systems from A Single Trajectory

Navid Reyhanian, Jarvis Haupt

This work investigates the problem of estimating the weight matrices of a stable time-invariant linear dynamical system from a single sequence of noisy measurements. We show that i…

cs.LG2020

Convexifying Sparse Interpolation with Infinitely Wide Neural Networks: An Atomic Norm Approach

Akshay Kumar, Jarvis Haupt

This work examines the problem of exact data interpolation via sparse (neuron count), infinitely wide, single hidden layer neural networks with leaky rectified linear unit activati…

cs.LG20208 cited

Provable Online CP/PARAFAC Decomposition of a Structured Tensor via Dictionary Learning

Sirisha Rambhatla, Xingguo Li, Jarvis Haupt

We consider the problem of factorizing a structured 3-way tensor into its constituent Canonical Polyadic (CP) factors. This decomposition, which can be viewed as a generalization o…

math.OC2019

A Provably Communication-Efficient Asynchronous Distributed Inference Method for Convex and Nonconvex Problems

Jineng Ren, Jarvis Haupt

This paper proposes and analyzes a communication-efficient distributed optimization framework for general nonconvex nonsmooth signal processing and machine learning problems under…

cs.CV20195 cited

Target-based Hyperspectral Demixing via Generalized Robust PCA

Sirisha Rambhatla, Xingguo Li, Jarvis Haupt

Localizing targets of interest in a given hyperspectral (HS) image has applications ranging from remote sensing to surveillance. This task of target detection leverages the fact th…

cs.CV2019

A Dictionary-Based Generalization of Robust PCA Part II: Applications to Hyperspectral Demixing

Sirisha Rambhatla, Xingguo Li, Jineng Ren +1

We consider the task of localizing targets of interest in a hyperspectral (HS) image based on their spectral signature(s), by posing the problem as two distinct convex demixing tas…