Connections between sparse estimation and robust statistical learning

Tsakonas, Efthymios; Jalden, J.; Sidiropoulos, N.D.; Ottersten, Björn

doi:10.1109/ICASSP.2013.6638713

Reference : Connections between sparse estimation and robust statistical learning


	Document type :	Scientific congresses, symposiums and conference proceedings : Paper published in a book
	Discipline(s) :	Engineering, computing & technology : Computer science
	To cite this reference:	http://hdl.handle.net/10993/14895

Title :	Connections between sparse estimation and robust statistical learning
Language :	English
Author, co-author :	Tsakonas, Efthymios [> >]
	Jalden, J. [> >]
	Sidiropoulos, N.D. [> >]
	Ottersten, Björn [University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > >]
Publication date :	2013
Main document title :	Acoustics, Speech and Signal Processing (ICASSP), 2013 IEEE International Conference on
Peer reviewed :	Yes
Audience :	International
Event name :	Acoustics, Speech and Signal Processing (ICASSP), 2013 IEEE International Conference on
Event date :	from 26-05-2013 to 31-05-2013
Event organizer :	IEEE
Event place (city) :	Vancouver
Event country :	Canada
Abstract :	[en] Recent literature on robust statistical inference suggests that promising outlier rejection schemes can be based on accounting explicitly for sparse gross errors in the modeling, and then relying on compressed sensing ideas to perform the outlier detection. In this paper, we consider two models for recovering a sparse signal from noisy measurements, possibly also contaminated with outliers. The models considered here are a linear regression model, and its natural one-bit counterpart where measurements are additionally quantized to a single bit. Our contributions can be summarized as follows: We start by providing conditions for identification and the Cramér-Rao Lower Bounds (CRLBs) for these two models. Then, focusing on the one-bit model, we derive conditions for consistency of the associated Maximum Likelihood estimator, and show the performance of relevant ℓ1-based relaxation strategies by comparing against the theoretical CRLB.
Target :	Researchers ; Professionals ; Students ; Others
Permalink :	http://hdl.handle.net/10993/14895
DOI :	10.1109/ICASSP.2013.6638713

File(s) associated to this reference

Fulltext file(s):

	File	Commentary	Version	Size	Access
Limited access	Connections between sparse estimation and robust statistical learning.pdf		Publisher postprint	186.75 kB	Request a copy

All documents in ORBi^lu are protected by a user license.

University of Luxembourg Library | Feedback | Legal notices