Cost-Sensitive Learning of Deep Feature Representations From Imbalanced Data

Salman H. Khan, Munawar Hayat, Mohammed Bennamoun, Ferdous A. Sohel, Roberto Togneri

Research output: Contribution to journalArticle

43 Citations (Scopus)

Abstract

Class imbalance is a common problem in the case of real-world object detection and classification tasks. Data of some classes are abundant, making them an overrepresented majority, and data of other classes are scarce, making them an underrepresented minority. This imbalance makes it challenging for a classifier to appropriately learn the discriminating boundaries of the majority and minority classes. In this paper, we propose a cost-sensitive (CoSen) deep neural network, which can automatically learn robust feature representations for both the majority and minority classes. During training, our learning procedure jointly optimizes the class-dependent costs and the neural network parameters. The proposed approach is applicable to both binary and multiclass problems without any modification. Moreover, as opposed to data-level approaches, we do not alter the original data distribution, which results in a lower computational cost during the training process. We report the results of our experiments on six major image classification data sets and show that the proposed approach significantly outperforms the baseline algorithms. Comparisons with popular data sampling techniques and CoSen classifiers demonstrate the superior performance of our proposed method.

Original languageEnglish
Pages (from-to)3573-3587
JournalIEEE Transactions on Neural Networks and Learning Systems
Volume29
Issue number8
DOIs
Publication statusPublished - 17 Aug 2017

Fingerprint

Costs
Classifiers
Image classification
Sampling
Neural networks
Experiments
Object detection
Deep neural networks

Cite this

@article{c7d04e2844ab4fbca8b3dee336adabf5,
title = "Cost-Sensitive Learning of Deep Feature Representations From Imbalanced Data",
abstract = "Class imbalance is a common problem in the case of real-world object detection and classification tasks. Data of some classes are abundant, making them an overrepresented majority, and data of other classes are scarce, making them an underrepresented minority. This imbalance makes it challenging for a classifier to appropriately learn the discriminating boundaries of the majority and minority classes. In this paper, we propose a cost-sensitive (CoSen) deep neural network, which can automatically learn robust feature representations for both the majority and minority classes. During training, our learning procedure jointly optimizes the class-dependent costs and the neural network parameters. The proposed approach is applicable to both binary and multiclass problems without any modification. Moreover, as opposed to data-level approaches, we do not alter the original data distribution, which results in a lower computational cost during the training process. We report the results of our experiments on six major image classification data sets and show that the proposed approach significantly outperforms the baseline algorithms. Comparisons with popular data sampling techniques and CoSen classifiers demonstrate the superior performance of our proposed method.",
keywords = "Australia, Computer vision, Convolutional neural networks (CNNs), cost-sensitive (CoSen) learning, data imbalance, loss functions., Neural networks, Tag clouds, Testing, Training, Training data",
author = "Khan, {Salman H.} and Munawar Hayat and Mohammed Bennamoun and Sohel, {Ferdous A.} and Roberto Togneri",
year = "2017",
month = "8",
day = "17",
doi = "10.1109/TNNLS.2017.2732482",
language = "English",
volume = "29",
pages = "3573--3587",
journal = "IEEE Transactions on Neural Networks",
issn = "1045-9227",
publisher = "IEEE, Institute of Electrical and Electronics Engineers",
number = "8",

}

Cost-Sensitive Learning of Deep Feature Representations From Imbalanced Data. / Khan, Salman H.; Hayat, Munawar; Bennamoun, Mohammed; Sohel, Ferdous A.; Togneri, Roberto.

In: IEEE Transactions on Neural Networks and Learning Systems, Vol. 29, No. 8, 17.08.2017, p. 3573-3587.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Cost-Sensitive Learning of Deep Feature Representations From Imbalanced Data

AU - Khan, Salman H.

AU - Hayat, Munawar

AU - Bennamoun, Mohammed

AU - Sohel, Ferdous A.

AU - Togneri, Roberto

PY - 2017/8/17

Y1 - 2017/8/17

N2 - Class imbalance is a common problem in the case of real-world object detection and classification tasks. Data of some classes are abundant, making them an overrepresented majority, and data of other classes are scarce, making them an underrepresented minority. This imbalance makes it challenging for a classifier to appropriately learn the discriminating boundaries of the majority and minority classes. In this paper, we propose a cost-sensitive (CoSen) deep neural network, which can automatically learn robust feature representations for both the majority and minority classes. During training, our learning procedure jointly optimizes the class-dependent costs and the neural network parameters. The proposed approach is applicable to both binary and multiclass problems without any modification. Moreover, as opposed to data-level approaches, we do not alter the original data distribution, which results in a lower computational cost during the training process. We report the results of our experiments on six major image classification data sets and show that the proposed approach significantly outperforms the baseline algorithms. Comparisons with popular data sampling techniques and CoSen classifiers demonstrate the superior performance of our proposed method.

AB - Class imbalance is a common problem in the case of real-world object detection and classification tasks. Data of some classes are abundant, making them an overrepresented majority, and data of other classes are scarce, making them an underrepresented minority. This imbalance makes it challenging for a classifier to appropriately learn the discriminating boundaries of the majority and minority classes. In this paper, we propose a cost-sensitive (CoSen) deep neural network, which can automatically learn robust feature representations for both the majority and minority classes. During training, our learning procedure jointly optimizes the class-dependent costs and the neural network parameters. The proposed approach is applicable to both binary and multiclass problems without any modification. Moreover, as opposed to data-level approaches, we do not alter the original data distribution, which results in a lower computational cost during the training process. We report the results of our experiments on six major image classification data sets and show that the proposed approach significantly outperforms the baseline algorithms. Comparisons with popular data sampling techniques and CoSen classifiers demonstrate the superior performance of our proposed method.

KW - Australia

KW - Computer vision

KW - Convolutional neural networks (CNNs)

KW - cost-sensitive (CoSen) learning

KW - data imbalance

KW - loss functions.

KW - Neural networks

KW - Tag clouds

KW - Testing

KW - Training

KW - Training data

UR - http://www.scopus.com/inward/record.url?scp=85028455314&partnerID=8YFLogxK

U2 - 10.1109/TNNLS.2017.2732482

DO - 10.1109/TNNLS.2017.2732482

M3 - Article

VL - 29

SP - 3573

EP - 3587

JO - IEEE Transactions on Neural Networks

JF - IEEE Transactions on Neural Networks

SN - 1045-9227

IS - 8

ER -