Optimizing quantization for Lasso recovery

Authors

Xiaoyi Gu, University of California, Los Angeles
Shenyinying Tu, University of California, Los Angeles
Hao-Jun Michael Shi, University of California, Los Angeles
Mindy Case, University of California, Los Angeles
Deanna Needell, Claremont McKenna CollegeFollow
Yaniv Plan, University of British Columbia

Document Type

Article - preprint

Department

Claremont McKenna College, Mathematics (CMC)

Publication Date

6-10-2016

Abstract

This letter is focused on quantized Compressed Sensing, assuming that Lasso is used for signal estimation. Leveraging recent work, we provide a framework to optimize the quantization function and show that the recovered signal converges to the actual signal at a quadratic rate as a function of the quantization level. We show that when the number of observations is high, this method of quantization gives a significantly better recovery rate than standard Lloyd-Max quantization. We support our theoretical analysis with numerical simulations.

Rights Information

© 2016 Gu, Tu, Shi, Case, Needell, Plan

Terms of Use & License Information

Terms of Use for work posted in Scholarship@Claremont.

Recommended Citation

X. Gu, S. Tu, H-J.M. Shi, M. Case, D. Needell, and Y. Plan. “Optimizing quantization for Lasso recovery.” 2016.