Please use this identifier to cite or link to this item:
https://hdl.handle.net/2440/78929
Citations | ||
Scopus | Web of Science® | Altmetric |
---|---|---|
?
|
?
|
Type: | Journal article |
Title: | Fully corrective boosting with arbitrary loss and regularization |
Author: | Shen, C. Li, H. Van Den Hengel, A. |
Citation: | Neural Networks, 2013; 48:44-58 |
Publisher: | Pergamon-Elsevier Science Ltd |
Issue Date: | 2013 |
ISSN: | 0893-6080 1879-2782 |
Statement of Responsibility: | Chunhua Shen, Hanxi Li, Anton van den Hengel |
Abstract: | We propose a general framework for analyzing and developing fully corrective boosting-based classifiers. The framework accepts any convex objective function, and allows any convex (for example, lp-norm, p ≥ 1) regularization term. By placing the wide variety of existing fully corrective boosting-based classifiers on a common footing, and considering the primal and dual problems together, the framework allows direct com- parison between apparently disparate methods. By solving the primal rather than the dual the framework is capable of generating efficient fully-corrective boosting algorithms without recourse to sophisticated convex optimization processes. We show that a range of additional boosting-based algorithms can be incorporated into the framework despite not being fully corrective. Finally, we provide an empirical analysis of the per- formance of a variety of the most significant boosting-based classifiers on a few machine learning benchmark datasets. |
Keywords: | Boosting ensemble learning convex optimization column generation |
Rights: | Copyright © 2013 Elsevier Ltd. |
DOI: | 10.1016/j.neunet.2013.07.006 |
Grant ID: | http://purl.org/au-research/grants/arc/FT120100969 http://purl.org/au-research/grants/arc/FT120100969 |
Published version: | http://dx.doi.org/10.1016/j.neunet.2013.07.006 |
Appears in Collections: | Aurora harvest Computer Science publications |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
hdl_78929.pdf | Accepted version | 549.63 kB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.