TaylorBoost: First and second-order boosting algorithms with explicit margin control

Authors:
M. J. Saberian;H. Masnadi-Shirazi;N. Vasconcelos
Affiliations:
Dept. of Electr. & Comput. Eng., Univ. of California, San Diego, CA, USA;Dept. of Electr. & Comput. Eng., Univ. of California, San Diego, CA, USA;Dept. of Electr. & Comput. Eng., Univ. of California, San Diego, CA, USA
Venue:
CVPR '11 Proceedings of the 2011 IEEE Conference on Computer Vision and Pattern Recognition
Year:
2011

Citing 0
Cited 3

Classifier-specific intermediate representation for multimedia tasks

Proceedings of the 2nd ACM International Conference on Multimedia Retrieval
Knowledge adaptation for ad hoc multimedia event detection with few exemplars

Proceedings of the 20th ACM international conference on Multimedia
Thinking of images as what they are: compound matrix regression for image classification

IJCAI'13 Proceedings of the Twenty-Third international joint conference on Artificial Intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

A new family of boosting algorithms, denoted Taylor-Boost, is proposed. It supports any combination of loss function and first or second order optimization, and includes classical algorithms such as AdaBoost, Gradient-Boost, or LogitBoost as special cases. Its restriction to the set of canonical losses makes it possible to have boosting algorithms with explicit margin control. A new large family of losses with this property, based on the set of cumulative distributions of zero mean random variables, is then proposed. A novel loss function in this family, the Laplace loss, is finally derived. The combination of this loss and second order TaylorBoost produces a boosting algorithm with explicit margin control.