Feature Selection for SVMs
We introduce a method of feature selection for Support Vector Machines. The method is based upon finding those features which minimize bounds on the leave-one-out error. This search can be efficiently performed via gradient descent. The resulting algorithms are shown to be superior to some standard...
| Main Authors: | , , , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | en_US |
| Published: |
Neural Information Processing Systems Foundation
2016
|
| Online Access: | http://hdl.handle.net/1721.1/102484 https://orcid.org/0000-0002-3944-0455 |
| _version_ | 1811079922169413632 |
|---|---|
| author | Poggio, Tomaso A. Weston, Jason Mukherjee, Sayan Pontil, Massimiliano Chapelle, Olivier Vapnik, Vladimir |
| author2 | Massachusetts Institute of Technology. Center for Biological & Computational Learning |
| author_facet | Massachusetts Institute of Technology. Center for Biological & Computational Learning Poggio, Tomaso A. Weston, Jason Mukherjee, Sayan Pontil, Massimiliano Chapelle, Olivier Vapnik, Vladimir |
| author_sort | Poggio, Tomaso A. |
| collection | MIT |
| description | We introduce a method of feature selection for Support Vector Machines. The method is based upon finding those features which minimize bounds on the leave-one-out error. This search can be efficiently performed via gradient descent. The resulting algorithms are shown to be superior to some standard feature selection algorithms on both toy data and real-life problems of face recognition, pedestrian detection and analyzing DNA micro array data. |
| first_indexed | 2024-09-23T11:22:36Z |
| format | Article |
| id | mit-1721.1/102484 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T11:22:36Z |
| publishDate | 2016 |
| publisher | Neural Information Processing Systems Foundation |
| record_format | dspace |
| spelling | mit-1721.1/1024842022-10-01T03:13:14Z Feature Selection for SVMs Poggio, Tomaso A. Weston, Jason Mukherjee, Sayan Pontil, Massimiliano Chapelle, Olivier Vapnik, Vladimir Massachusetts Institute of Technology. Center for Biological & Computational Learning Massachusetts Institute of Technology. Department of Brain and Cognitive Sciences Poggio, Tomaso A. Mukherjee, Sayan Pontil, Massimiliano We introduce a method of feature selection for Support Vector Machines. The method is based upon finding those features which minimize bounds on the leave-one-out error. This search can be efficiently performed via gradient descent. The resulting algorithms are shown to be superior to some standard feature selection algorithms on both toy data and real-life problems of face recognition, pedestrian detection and analyzing DNA micro array data. 2016-05-13T18:41:58Z 2016-05-13T18:41:58Z 2000 Article http://purl.org/eprint/type/ConferencePaper http://hdl.handle.net/1721.1/102484 Weston, J., S. Mukherjee, O. Chapelle, M. Pontil, T. Poggio, and V. Vapnik. "Feature Selection for SVMs." Advances in Neural Information Processing Systems 13 (NIPS 2000). © 2000 Neural Information Processing Systems Foundation, Inc. https://orcid.org/0000-0002-3944-0455 en_US https://papers.nips.cc/paper/1850-feature-selection-for-svms Advances in Neural Information Processing Systems (NIPS) Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use. application/pdf Neural Information Processing Systems Foundation NIPS |
| spellingShingle | Poggio, Tomaso A. Weston, Jason Mukherjee, Sayan Pontil, Massimiliano Chapelle, Olivier Vapnik, Vladimir Feature Selection for SVMs |
| title | Feature Selection for SVMs |
| title_full | Feature Selection for SVMs |
| title_fullStr | Feature Selection for SVMs |
| title_full_unstemmed | Feature Selection for SVMs |
| title_short | Feature Selection for SVMs |
| title_sort | feature selection for svms |
| url | http://hdl.handle.net/1721.1/102484 https://orcid.org/0000-0002-3944-0455 |
| work_keys_str_mv | AT poggiotomasoa featureselectionforsvms AT westonjason featureselectionforsvms AT mukherjeesayan featureselectionforsvms AT pontilmassimiliano featureselectionforsvms AT chapelleolivier featureselectionforsvms AT vapnikvladimir featureselectionforsvms |