Parallelizing Basis Pursuit Denoising

Cory Kromer-Edwards; Suely Oliveira; David Stewart

doi:10.1109/IJCNN.2019.8851987

Back

Conference proceeding

Parallelizing Basis Pursuit Denoising

Cory Kromer-Edwards, Suely Oliveira and David Stewart

2019 International Joint Conference on Neural Networks (IJCNN), Vol.2019-, pp.1-5

07/2019

DOI: 10.1109/IJCNN.2019.8851987

View Online

Abstract

Basis Pursuit DeNoising (BPDN) is a version of sparse least squares, where there is an ℓ 1 penalty for the coefficients in addition to the sum of squares of the errors. To apply this technique to large data sets, parallel algorithms are needed. In this paper, we create a parallel version of a BPDN algorithm [11] that can compute exact solutions for an interval of values for the ℓ 1 penalty parameter that exploits the sparsity of the solutions, unlike stochastic gradient descent-type algorithms.

Artificial intelligence

Computer science

Neural networks

Noise reduction

Parallel algorithms

Testing

Urban areas

Details

Title: Subtitle: Parallelizing Basis Pursuit Denoising
Creators: Cory Kromer-Edwards - University of Iowa
Suely Oliveira - University of Iowa
David Stewart - University of Iowa
Resource Type: Conference proceeding
Publication Details: 2019 International Joint Conference on Neural Networks (IJCNN), Vol.2019-, pp.1-5
DOI: 10.1109/IJCNN.2019.8851987
eISSN: 2161-4407
Publisher: IEEE
Language: English
Date published: 07/2019
Academic Unit: Computer Science; Mathematics
Record Identifier: 9984241051702771

Metrics

9 Record Views