ARConvL: Adaptive Region-Based Convolutional Learning for Multi-class Imbalance Classification

Shuxian LI, Liyan SONG*, Xiaoyu WU, Zheng HU, Yiu-ming CHEUNG, Xin YAO*

*Corresponding author for this work

Research output: Book Chapters | Papers in Conference ProceedingsConference paper (refereed)Researchpeer-review

Abstract

Real-world image classification usually suffers from the multi-class imbalance issue, probably causing unsatisfactory performance, especially on minority classes. A typical way to address such problem is to adjust the loss function of deep networks by making use of class imbalance ratios. However, such static between-class imbalance ratios cannot monitor the changing latent feature distributions that are continuously learned by the deep network throughout training epochs, potentially failing in helping the loss function adapt to the latest class imbalance status of the current training epoch. To address this issue, we propose an adaptive loss to monitor the evolving learning of latent feature distributions. Specifically, the class-wise feature distribution is derived based on the region loss with the objective of accommodating feature points of this class. The multi-class imbalance issue can then be addressed based on the derived class regions from two perspectives: first, an adaptive distribution loss is proposed to optimize class-wise latent feature distributions where different classes would converge within the regions of a similar size, directly tackling the multi-class imbalance problem; second, an adaptive margin is proposed to incorporate with the cross-entropy loss to enlarge the between-class discrimination, further alleviating the class imbalance issue. An adaptive region-based convolutional learning method is ultimately produced based on the adaptive distribution loss and the adaptive margin cross-entropy loss. Experimental results based on public image sets demonstrate the effectiveness and robustness of our approach in dealing with varying levels of multi-class imbalance issues. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.
Original languageEnglish
Title of host publicationMachine Learning and Knowledge Discovery in Databases : Research Track : European Conference, ECML PKDD 2023, Turin, Italy, September 18–22, 2023, Proceedings, Part II
EditorsDanai KOUTRA, Claudia PLANT, Manuel Gomez RODRIGUEZ, Elena BARALIS, Francesco BONCHI
PublisherSpringer Science and Business Media Deutschland GmbH
Pages103-120
Number of pages18
ISBN (Electronic)9783031434150
ISBN (Print)9783031434143
DOIs
Publication statusPublished - 2023
Externally publishedYes
EventThe 2023 edition of the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases - Turin, Italy
Duration: 18 Sept 202322 Sept 2023

Publication series

NameLecture Notes in Computer Science
PublisherSpringer
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349
NameLecture Notes in Artificial Intelligence
PublisherSpringer
Volume14170
ISSN (Print)2945-9133
ISSN (Electronic)2945-9141

Conference

ConferenceThe 2023 edition of the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases
Abbreviated titleECML PKDD 2023
Country/TerritoryItaly
CityTurin
Period18/09/2322/09/23

Bibliographical note

This work was supported by National Natural Science Foundation of China (NSFC) under Grant No. 62002148 and Grant No. 62250710682, Guangdong Provincial Key Laboratory under Grant No. 2020B121201001, the Program for Guangdong Introducing Innovative and Enterpreneurial Teams under Grant No. 2017ZT07X386, and Research Institute of Trustworthy Autonomous Systems (RITAS).

Keywords

  • Adaptive loss
  • Deep learning
  • Feature engineering
  • Margin
  • Multi-class imbalance classification

Fingerprint

Dive into the research topics of 'ARConvL: Adaptive Region-Based Convolutional Learning for Multi-class Imbalance Classification'. Together they form a unique fingerprint.

Cite this