Conference proceeding
Overlap Group Lasso for Variable Selection in Nonlinear Nonparametric System Identification
Proceedings of the 6th International Conference of Control, Dynamic Systems, and Robotics (CDSR'19)
International Conference of Control, Dynamic Systems, and Robotics, 6th (Ottawa, Canada, 06/06/2019 - 06/07/2019)
2019
DOI: 10.1159/cdsr19.137
Abstract
Identification of a nonlinear nonparametric system is not easy. On the other hand, many systems are sparse in the sense that not all variables contribute. If these variables that do not contribute can be detected and removed, the identification problem becomes lower dimensional and is relatively easy to deal with. The goal of the paper is to develop an overlap group Lasso method to detect which variables contribute and which variables do not. The algorithm developed favors sparsity in terms of partial derivatives that is the necessary and sufficient condition for a variable to contribute.
Details
- Title: Subtitle
- Overlap Group Lasso for Variable Selection in Nonlinear Nonparametric System Identification
- Creators
- Changming Cheng (Author) - Shanghai Jiao Tong UniversityEr-Wei Bai (Author) - University of Iowa, Electrical and Computer Engineering
- Resource Type
- Conference proceeding
- Publication Details
- Proceedings of the 6th International Conference of Control, Dynamic Systems, and Robotics (CDSR'19)
- Conference
- International Conference of Control, Dynamic Systems, and Robotics, 6th (Ottawa, Canada, 06/06/2019 - 06/07/2019)
- DOI
- 10.1159/cdsr19.137
- Language
- English
- Date published
- 2019
- Academic Unit
- Electrical and Computer Engineering
- Record Identifier
- 9984582847802771
Metrics
1 Record Views