Revisiting Model Selection and Recovery of Sparse Signals Using One-Step Thresholding

This paper studies non-asymptotic model selection and recovery of sparse signals in high-dimensional, linear inference problems. In contrast to the existing literature, the focus here is on the general case of arbitrary design matrices and arbitrary nonzero entries of the signal. In this regard, it utilizes two easily computable measures of coherence - termed as the worst-case coherence and the average coherence - among the columns of a design matrix to analyze a simple, model-order agnostic One-Step Thresholding (OST) algorithm.

Provided by: Princeton Software Topic: Networking Date Added: Oct 2010 Format: PDF

Find By Topic