Saved in:
Bibliographic Details
Main Authors: Fernández-Baldera, Antonio, Buenaposada, José M., Baumela, Luis
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2402.04465
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916117538144256
author Fernández-Baldera, Antonio
Buenaposada, José M.
Baumela, Luis
author_facet Fernández-Baldera, Antonio
Buenaposada, José M.
Baumela, Luis
contents We present BAdaCost, a multi-class cost-sensitive classification algorithm. It combines a set of cost-sensitive multi-class weak learners to obtain a strong classification rule within the Boosting framework. To derive the algorithm we introduce CMEL, a Cost-sensitive Multi-class Exponential Loss that generalizes the losses optimized in various classification algorithms such as AdaBoost, SAMME, Cost-sensitive AdaBoost and PIBoost. Hence unifying them under a common theoretical framework. In the experiments performed we prove that BAdaCost achieves significant gains in performance when compared to previous multi-class cost-sensitive approaches. The advantages of the proposed algorithm in asymmetric multi-class classification are also evaluated in practical multi-view face and car detection problems.
format Preprint
id arxiv_https___arxiv_org_abs_2402_04465
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle BAdaCost: Multi-class Boosting with Costs
Fernández-Baldera, Antonio
Buenaposada, José M.
Baumela, Luis
Computer Vision and Pattern Recognition
We present BAdaCost, a multi-class cost-sensitive classification algorithm. It combines a set of cost-sensitive multi-class weak learners to obtain a strong classification rule within the Boosting framework. To derive the algorithm we introduce CMEL, a Cost-sensitive Multi-class Exponential Loss that generalizes the losses optimized in various classification algorithms such as AdaBoost, SAMME, Cost-sensitive AdaBoost and PIBoost. Hence unifying them under a common theoretical framework. In the experiments performed we prove that BAdaCost achieves significant gains in performance when compared to previous multi-class cost-sensitive approaches. The advantages of the proposed algorithm in asymmetric multi-class classification are also evaluated in practical multi-view face and car detection problems.
title BAdaCost: Multi-class Boosting with Costs
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2402.04465