Cost-sensitive boosting algorithms as gradient descent

AdaBoost is a well known boosting method for generating strong ensemble of weak base learners. The procedure of AdaBoost can be fitted in a gradient descent optimization framework, which is important for analyzing and devising its procedure. Cost sensitive boosting (CSB) is an emerging subject exten...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2008 IEEE International Conference on Acoustics, Speech and Signal Processing s. 2009 - 2012
Hlavní autoři: Qu-Tang Cai, Yang-Qui Song, Chang-Shui Zhang
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 01.03.2008
Témata:
Algorithm design and analysis
Automation
Boosting
Classification algorithms
Cost-sensitive Classification
Costs
Gradient Descent
Information science
Intelligent systems
Laboratories
Optimization
Pattern recognition
ISBN:9781424414833, 1424414830
ISSN:1520-6149
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:AdaBoost is a well known boosting method for generating strong ensemble of weak base learners. The procedure of AdaBoost can be fitted in a gradient descent optimization framework, which is important for analyzing and devising its procedure. Cost sensitive boosting (CSB) is an emerging subject extending the boosting methods for cost sensitive classification applications. Most CSB methods are performed by directly modifying the original AdaBoost procedure. Unfortunately, the effectiveness of most cost sensitive boosting methods are checked only by experiments. It remains unclear whether these methods can be viewed as gradient descent procedures like AdaBoost. In this paper, we show that several typical CSB methods can also be view as gradient descent for minimizing a unified objective function. We then deduce a general greedy boosting procedure. Experimental results also validate the effectiveness of the proposed procedure.
ISBN:9781424414833
1424414830
ISSN:1520-6149
DOI:10.1109/ICASSP.2008.4518033