Compressed learning: Universal sparse dimensionality reduction and learning in the measurement domain

Compressed learning: Universal sparse dimensionality reduction and learning in the measurement domain (2009)

Cached

Download Links

by Robert Calderbank , Sina Jafarpour , Robert Schapire

Citations:

7 - 0 self

Versions

Version History

Metadata Version 1

Datum	Value	Source
TITLE	Compressed learning: Universal sparse dimensionality reduction and learning in the measurement domain	INFERENCE
AUTHOR NAME	Robert Calderbank	SVM HeaderParse 0.2
AUTHOR AFFIL	Electrical Engineering and Mathematics; Princeton University	SVM HeaderParse 0.2
AUTHOR NAME	Sina Jafarpour	SVM HeaderParse 0.2
AUTHOR AFFIL	Computer Science; Princeton University	SVM HeaderParse 0.2
AUTHOR NAME	Robert Schapire	SVM HeaderParse 0.2
AUTHOR AFFIL	Computer Science; Princeton University	SVM HeaderParse 0.2
ABSTRACT	In this paper, we provide theoretical results to show that compressed learning, learning directly in the compressed domain, is possible. In Particular, we provide tight bounds demonstrating that the linear kernel SVM’s classifier in the measurement domain, with high probability, has true accuracy close to the accuracy of the best linear threshold classifier in the data domain. We show that this is beneficial both from the compressed sensing and the machine learning points of view. Furthermore, we indicate that for a family of well-known compressed sensing matrices, compressed learning is universal, in the sense that learning and classification in the measurement domain works provided that the data are sparse in some, even unknown, basis. Moreover, we show that our results are also applicable to a family of smooth manifold-learning tasks. Finally, we support our claims with experimental results. 1	SVM HeaderParse 0.2
YEAR	2009	INFERENCE
VENUE TYPE	TECHREPORT	INFERENCE
TECH	Technical Report	INFERENCE
CITATIONS	38 found	ParsCit 1.0