The group-lasso : two novel applications
Abstract/Contents
- Abstract
- In the first application, we introduce a method for learning pairwise interactions in a manner that satisfies strong hierarchy: whenever an interaction is estimated to be nonzero, both its associated main effects are also included in the model. We compare our method with existing approaches on both simulated and real data, including a genome wide association study, all using our R package glinternet. The second application is about recovering neural source activity in the visual cortex using non-invasive electroencephalography (EEG) recordings from sensors placed around a subject's head. We show that the group-lasso outperforms the widely-used minimum norm inversion, and that the group-lasso performance improves with the number of subjects. We also show that averaging the estimated source activity within appropriately defined regions of interest (ROIs) in the visual cortex across multiple subjects is able to dramatically boost the performance of both the minimum norm and group-lasso solutions, and also improves with the number of subjects.
Description
Type of resource | text |
---|---|
Form | electronic; electronic resource; remote |
Extent | 1 online resource. |
Publication date | 2013 |
Issuance | monographic |
Language | English |
Creators/Contributors
Associated with | Lim, Michael |
---|---|
Associated with | Stanford University, Department of Statistics. |
Primary advisor | Hastie, Trevor |
Thesis advisor | Hastie, Trevor |
Thesis advisor | Taylor, Jonathan E |
Thesis advisor | Tibshirani, Robert |
Advisor | Taylor, Jonathan E |
Advisor | Tibshirani, Robert |
Subjects
Genre | Theses |
---|
Bibliographic information
Statement of responsibility | Michael Lim. |
---|---|
Note | Submitted to the Department of Statistics. |
Thesis | Thesis (Ph.D.)--Stanford University, 2013. |
Location | electronic resource |
Access conditions
- Copyright
- © 2013 by Michael Heng Keng Lim
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...