On semi-supervised learning and sparsity

Balinsky, Alexander ORCID: https://orcid.org/0000-0002-8151-4462 and Balinsky, Helen 2009. On semi-supervised learning and sparsity. Presented at: IEEE International Conference on Systems, Man, and Cybernetics, San Antonio, TX, USA, 11-14 October 2009. Proceedings 2009 International Conference on Systems, Man and Cybernetics October 11-14, 2009 : San Antonio, Texas, USA. Los Alamitos, CA: IEEE, pp. 3083-3087. 10.1109/ICSMC.2009.5345946

Full text not available from this repository.

Abstract

In this article we establish a connection between semi-supervised learning and compressive sampling. We show that sparsity and compressibility of the learning function can be obtained from heavy-tailed distributions of filter responses or coefficients in spectral decompositions. In many cases the NP-hard problems of finding sparsest solutions can be replaced by l1-problems from convex optimisation theory, which provide effective tools for semi-supervised learning. We present several conjectures and examples.

Item Type:	Conference or Workshop Item (Paper)
Date Type:	Publication
Status:	Published
Schools:	Mathematics
Subjects:	Q Science > QA Mathematics Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Uncontrolled Keywords:	Semi-supervised learning; compressive sampling; heavy-tailed distributions; sparsity
Publisher:	IEEE
ISBN:	9781424427932
Last Modified:	19 Oct 2022 10:28
URI:	https://orca.cardiff.ac.uk/id/eprint/24448

Actions (repository staff only)

Edit Item