An easy-to-hard learning paradigm for multiple classes and multiple labels

Weiwei Liu, Ivor W. Tsang, Klaus Muller

Research output: Contribution to journalArticle

18 Citations (Scopus)

Abstract

Many applications, such as human action recognition and object detection, can be formulated as a multiclass classification problem. One-vs-rest (OVR) is one of the most widely used approaches for multiclass classification due to its simplicity and excellent performance. However, many confusing classes in such applications will degrade its results. For example, hand clap and boxing are two confusing actions. Hand clap is easily misclassified as boxing, and vice versa. Therefore, precisely classifying confusing classes remains a challenging task. To obtain better performance for multiclass classifications that have confusing classes, we first develop a classifier chain model for multiclass classification (CCMC) to transfer class information between classifiers. Then, based on an analysis of our proposed model, we propose an easy-to-hard learning paradigm for multiclass classification to automatically identify easy and hard classes and then use the predictions from simpler classes to help solve harder classes. Similar to CCMC, the classifier chain (CC) model is also proposed by Read et al. (2009) to capture the label dependency for multi-label classification. However, CC does not consider the order of difficulty of the labels and achieves degenerated performance when there are many confusing labels. Therefore, it is non-trivial to learn the appropriate label order for CC. Motivated by our analysis for CCMC, we also propose the easy-to-hard learning paradigm for multi-label classi cation to automatically identify easy and hard labels, and then use the predictions from simpler labels to help solve harder labels. We also demonstrate that our proposed strategy can be successfully applied to a wide range of applications, such as ordinal classi cation and relationship prediction. Extensive empirical studies validate our analysis and the e-ectiveness of our proposed easy-to-hard learning strategies.

Original languageEnglish
Pages (from-to)1-38
Number of pages38
JournalJournal of Machine Learning Research
Volume18
Publication statusPublished - 2017 Sep 1

Fingerprint

Multi-class Classification
Labels
Classifier
Paradigm
Classifiers
Prediction
Action Recognition
Model
Learning Strategies
Positive ions
Object Detection
Class
Learning
Classification Problems
Empirical Study
Simplicity
Range of data
Demonstrate

Keywords

  • Classier Chain
  • Easyto-hard Learning Paradigm
  • Multi-label Classication
  • Multiclass Classication

ASJC Scopus subject areas

  • Software
  • Control and Systems Engineering
  • Statistics and Probability
  • Artificial Intelligence

Cite this

An easy-to-hard learning paradigm for multiple classes and multiple labels. / Liu, Weiwei; Tsang, Ivor W.; Muller, Klaus.

In: Journal of Machine Learning Research, Vol. 18, 01.09.2017, p. 1-38.

Research output: Contribution to journalArticle

Liu, Weiwei ; Tsang, Ivor W. ; Muller, Klaus. / An easy-to-hard learning paradigm for multiple classes and multiple labels. In: Journal of Machine Learning Research. 2017 ; Vol. 18. pp. 1-38.
@article{8578908651ac40b8b04bf8fc189fc8d7,
title = "An easy-to-hard learning paradigm for multiple classes and multiple labels",
abstract = "Many applications, such as human action recognition and object detection, can be formulated as a multiclass classification problem. One-vs-rest (OVR) is one of the most widely used approaches for multiclass classification due to its simplicity and excellent performance. However, many confusing classes in such applications will degrade its results. For example, hand clap and boxing are two confusing actions. Hand clap is easily misclassified as boxing, and vice versa. Therefore, precisely classifying confusing classes remains a challenging task. To obtain better performance for multiclass classifications that have confusing classes, we first develop a classifier chain model for multiclass classification (CCMC) to transfer class information between classifiers. Then, based on an analysis of our proposed model, we propose an easy-to-hard learning paradigm for multiclass classification to automatically identify easy and hard classes and then use the predictions from simpler classes to help solve harder classes. Similar to CCMC, the classifier chain (CC) model is also proposed by Read et al. (2009) to capture the label dependency for multi-label classification. However, CC does not consider the order of difficulty of the labels and achieves degenerated performance when there are many confusing labels. Therefore, it is non-trivial to learn the appropriate label order for CC. Motivated by our analysis for CCMC, we also propose the easy-to-hard learning paradigm for multi-label classi cation to automatically identify easy and hard labels, and then use the predictions from simpler labels to help solve harder labels. We also demonstrate that our proposed strategy can be successfully applied to a wide range of applications, such as ordinal classi cation and relationship prediction. Extensive empirical studies validate our analysis and the e-ectiveness of our proposed easy-to-hard learning strategies.",
keywords = "Classier Chain, Easyto-hard Learning Paradigm, Multi-label Classication, Multiclass Classication",
author = "Weiwei Liu and Tsang, {Ivor W.} and Klaus Muller",
year = "2017",
month = "9",
day = "1",
language = "English",
volume = "18",
pages = "1--38",
journal = "Journal of Machine Learning Research",
issn = "1532-4435",
publisher = "Microtome Publishing",

}

TY - JOUR

T1 - An easy-to-hard learning paradigm for multiple classes and multiple labels

AU - Liu, Weiwei

AU - Tsang, Ivor W.

AU - Muller, Klaus

PY - 2017/9/1

Y1 - 2017/9/1

N2 - Many applications, such as human action recognition and object detection, can be formulated as a multiclass classification problem. One-vs-rest (OVR) is one of the most widely used approaches for multiclass classification due to its simplicity and excellent performance. However, many confusing classes in such applications will degrade its results. For example, hand clap and boxing are two confusing actions. Hand clap is easily misclassified as boxing, and vice versa. Therefore, precisely classifying confusing classes remains a challenging task. To obtain better performance for multiclass classifications that have confusing classes, we first develop a classifier chain model for multiclass classification (CCMC) to transfer class information between classifiers. Then, based on an analysis of our proposed model, we propose an easy-to-hard learning paradigm for multiclass classification to automatically identify easy and hard classes and then use the predictions from simpler classes to help solve harder classes. Similar to CCMC, the classifier chain (CC) model is also proposed by Read et al. (2009) to capture the label dependency for multi-label classification. However, CC does not consider the order of difficulty of the labels and achieves degenerated performance when there are many confusing labels. Therefore, it is non-trivial to learn the appropriate label order for CC. Motivated by our analysis for CCMC, we also propose the easy-to-hard learning paradigm for multi-label classi cation to automatically identify easy and hard labels, and then use the predictions from simpler labels to help solve harder labels. We also demonstrate that our proposed strategy can be successfully applied to a wide range of applications, such as ordinal classi cation and relationship prediction. Extensive empirical studies validate our analysis and the e-ectiveness of our proposed easy-to-hard learning strategies.

AB - Many applications, such as human action recognition and object detection, can be formulated as a multiclass classification problem. One-vs-rest (OVR) is one of the most widely used approaches for multiclass classification due to its simplicity and excellent performance. However, many confusing classes in such applications will degrade its results. For example, hand clap and boxing are two confusing actions. Hand clap is easily misclassified as boxing, and vice versa. Therefore, precisely classifying confusing classes remains a challenging task. To obtain better performance for multiclass classifications that have confusing classes, we first develop a classifier chain model for multiclass classification (CCMC) to transfer class information between classifiers. Then, based on an analysis of our proposed model, we propose an easy-to-hard learning paradigm for multiclass classification to automatically identify easy and hard classes and then use the predictions from simpler classes to help solve harder classes. Similar to CCMC, the classifier chain (CC) model is also proposed by Read et al. (2009) to capture the label dependency for multi-label classification. However, CC does not consider the order of difficulty of the labels and achieves degenerated performance when there are many confusing labels. Therefore, it is non-trivial to learn the appropriate label order for CC. Motivated by our analysis for CCMC, we also propose the easy-to-hard learning paradigm for multi-label classi cation to automatically identify easy and hard labels, and then use the predictions from simpler labels to help solve harder labels. We also demonstrate that our proposed strategy can be successfully applied to a wide range of applications, such as ordinal classi cation and relationship prediction. Extensive empirical studies validate our analysis and the e-ectiveness of our proposed easy-to-hard learning strategies.

KW - Classier Chain

KW - Easyto-hard Learning Paradigm

KW - Multi-label Classication

KW - Multiclass Classication

UR - http://www.scopus.com/inward/record.url?scp=85032904139&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85032904139&partnerID=8YFLogxK

M3 - Article

VL - 18

SP - 1

EP - 38

JO - Journal of Machine Learning Research

JF - Journal of Machine Learning Research

SN - 1532-4435

ER -