A Formulation for Active Learning with Applications to Object Detection
We discuss a formulation for active example selection for function learning problems. This formulation is obtained by adapting Fedorov's optimal experiment design to the learning problem. We specifically show how to analytically derive example selection algorithms for certain well defined...
| Main Authors: | , |
|---|---|
| Language: | en_US |
| Published: |
2004
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/1721.1/7209 |
| _version_ | 1826197524825767936 |
|---|---|
| author | Sung, Kah Kay Niyogi, Partha |
| author_facet | Sung, Kah Kay Niyogi, Partha |
| author_sort | Sung, Kah Kay |
| collection | MIT |
| description | We discuss a formulation for active example selection for function learning problems. This formulation is obtained by adapting Fedorov's optimal experiment design to the learning problem. We specifically show how to analytically derive example selection algorithms for certain well defined function classes. We then explore the behavior and sample complexity of such active learning algorithms. Finally, we view object detection as a special case of function learning and show how our formulation reduces to a useful heuristic to choose examples to reduce the generalization error. |
| first_indexed | 2024-09-23T10:49:19Z |
| id | mit-1721.1/7209 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T10:49:19Z |
| publishDate | 2004 |
| record_format | dspace |
| spelling | mit-1721.1/72092019-04-10T11:52:45Z A Formulation for Active Learning with Applications to Object Detection Sung, Kah Kay Niyogi, Partha active learning optimal experiment design object detection example selection We discuss a formulation for active example selection for function learning problems. This formulation is obtained by adapting Fedorov's optimal experiment design to the learning problem. We specifically show how to analytically derive example selection algorithms for certain well defined function classes. We then explore the behavior and sample complexity of such active learning algorithms. Finally, we view object detection as a special case of function learning and show how our formulation reduces to a useful heuristic to choose examples to reduce the generalization error. 2004-10-20T20:49:52Z 2004-10-20T20:49:52Z 1996-06-06 AIM-1438 CBCL-116 http://hdl.handle.net/1721.1/7209 en_US AIM-1438 CBCL-116 40 p. 593069 bytes 1090749 bytes application/octet-stream application/pdf application/octet-stream application/pdf |
| spellingShingle | active learning optimal experiment design object detection example selection Sung, Kah Kay Niyogi, Partha A Formulation for Active Learning with Applications to Object Detection |
| title | A Formulation for Active Learning with Applications to Object Detection |
| title_full | A Formulation for Active Learning with Applications to Object Detection |
| title_fullStr | A Formulation for Active Learning with Applications to Object Detection |
| title_full_unstemmed | A Formulation for Active Learning with Applications to Object Detection |
| title_short | A Formulation for Active Learning with Applications to Object Detection |
| title_sort | formulation for active learning with applications to object detection |
| topic | active learning optimal experiment design object detection example selection |
| url | http://hdl.handle.net/1721.1/7209 |
| work_keys_str_mv | AT sungkahkay aformulationforactivelearningwithapplicationstoobjectdetection AT niyogipartha aformulationforactivelearningwithapplicationstoobjectdetection AT sungkahkay formulationforactivelearningwithapplicationstoobjectdetection AT niyogipartha formulationforactivelearningwithapplicationstoobjectdetection |