Publication: Model and Feature Selection in Hidden Conditional Random Fields with Group Regularization
Loading...
Identifiers
ISSN: 0302-9743 (print)
ISSN: 1611-3349 (online)
ISBN: 978-3-642-40845-8 (print)
ISBN: 978-3-642-40846-5 (online)
UXXI: CC/0000021132
Publication date
2013
Defense date
Advisors
Tutors
Journal Title
Journal ISSN
Volume Title
Publisher
Springer
Impact
Export
Abstract
Sequence classification is an important problem in computer vision, speech analysis or computational biology. This paper presents a new training strategy for the Hidden Conditional Random Field sequence classifier incorporating model and feature selection. The standard Lasso regularization employed in the estimation of model parameters is replaced by overlapping group-L1 regularization. Depending on the configuration of the overlapping groups, model selection, feature selection,or both are performed. The sequence classifiers trained in this way have better predictive performance. The application of the proposed method in a human action recognition task confirms that fact.
Description
Proceedings of: 8th International Conference on Hybrid Artificial Intelligence Systems (HAIS 2013). Salamanca, September 11-13, 2013.
Keywords
Group Regularization, Human Action Recognition, HCRF, Hidden Conditional Random Field
Bibliographic citation
Pan, J. S. et al. (eds.) (2013). Hybrid Artificial Intelligent Systems: 8th International Conference, HAIS 2013, Salamanca, Spain, September 11-13, 2013. Proceedings. (Lecture Notes in Computer Science, 8073) Springer, 140-149.