31 citations · 186 across the 67 of their papers we have counts for
6 papers · 1 filter
Necessary and Sufficient Conditions for Novel Word Detection in Separable Topic Models
Weicong Ding, Prakash Ishwar, Mohammad H. Rohban +1
The simplicial condition and other stronger conditions that imply it have recently played a central role in developing polynomial time algorithms with provable asymptotic consisten…
Sparse Signal Recovery under Poisson Statistics
D. Motamedvaziri, M. H. Rohban, V. Saligrama
We are motivated by problems that arise in a number of applications such as Online Marketing and explosives detection, where the observations are usually modeled using Poisson stat…
Topic Discovery through Data Dependent and Random Projections
Weicong Ding, Mohammad H. Rohban, Prakash Ishwar +1
We present algorithms for topic modeling based on the geometry of cross-document word-frequency patterns. This perspective gains significance under the so called separability condi…
Incorporating Betweenness Centrality in Compressive Sensing for Congestion Detection
Hoda S. Ayatollahi Tabatabaii, Hamid R. Rabiee, Mohammad Hossein Rohban +1
This paper presents a new Compressive Sensing (CS) scheme for detecting network congested links. We focus on decreasing the required number of measurements to detect all congested…
A New Geometric Approach to Latent Topic Modeling and Discovery
Weicong Ding, Mohammad H. Rohban, Prakash Ishwar +1
A new geometrically-motivated algorithm for nonnegative matrix factorization is developed and applied to the discovery of latent "topics" for text and image "document" corpora. The…
An Impossibility Result for High Dimensional Supervised Learning
Mohammad Hossein Rohban, Prakash Ishwar, Birant Orten +2
We study high-dimensional asymptotic performance limits of binary supervised classification problems where the class conditional densities are Gaussian with unknown means and covar…