Conformity Region Classification with Instance-Transfer Boosting

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

Conformal region classification focuses on developing region classifiers; i.e., classifiers that output regions (sets) of classes for new test instances. 2,13,16 Conformal region classifiers have been proven to be valid for any significance level epsilon is an element of [0, 1] in the sense that the probability the class regions do not contain the true instances' classes does not exceed e. In practice, however, conformal region classifiers need to be also efficient; i.e., they have to output non-empty and relatively small class regions. In this paper we show that conformal region classification can benefit from instance-transfer learning. Our new approach consists of the basic conformal region classifier with a nonconformity function that implements instance transfer. We propose to learn such a function using a new multi-class Transfer AdaBoost. M1 algorithm. The function and its relation to the conformal region classification are theoretically justified. The experiments showed that our approach is valid for any significance level epsilon is an element of [0, 1] and its efficiency can be improved with instance transfer.
Original languageEnglish
JournalInternational Journal on Artificial Intelligence Tools
Volume24
Issue number6
DOIs
Publication statusPublished - 2015

Cite this

@article{002897b321ba4985b2cc1ec073059ede,
title = "Conformity Region Classification with Instance-Transfer Boosting",
abstract = "Conformal region classification focuses on developing region classifiers; i.e., classifiers that output regions (sets) of classes for new test instances. 2,13,16 Conformal region classifiers have been proven to be valid for any significance level epsilon is an element of [0, 1] in the sense that the probability the class regions do not contain the true instances' classes does not exceed e. In practice, however, conformal region classifiers need to be also efficient; i.e., they have to output non-empty and relatively small class regions. In this paper we show that conformal region classification can benefit from instance-transfer learning. Our new approach consists of the basic conformal region classifier with a nonconformity function that implements instance transfer. We propose to learn such a function using a new multi-class Transfer AdaBoost. M1 algorithm. The function and its relation to the conformal region classification are theoretically justified. The experiments showed that our approach is valid for any significance level epsilon is an element of [0, 1] and its efficiency can be improved with instance transfer.",
author = "S. Zhou and E.N. Smirnov and R. Peeters",
year = "2015",
doi = "10.1142/S0218213015600027",
language = "English",
volume = "24",
journal = "International Journal on Artificial Intelligence Tools",
issn = "0218-2130",
publisher = "World Scientific Publishing Company",
number = "6",

}

Conformity Region Classification with Instance-Transfer Boosting. / Zhou, S.; Smirnov, E.N.; Peeters, R.

In: International Journal on Artificial Intelligence Tools, Vol. 24, No. 6, 2015.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Conformity Region Classification with Instance-Transfer Boosting

AU - Zhou, S.

AU - Smirnov, E.N.

AU - Peeters, R.

PY - 2015

Y1 - 2015

N2 - Conformal region classification focuses on developing region classifiers; i.e., classifiers that output regions (sets) of classes for new test instances. 2,13,16 Conformal region classifiers have been proven to be valid for any significance level epsilon is an element of [0, 1] in the sense that the probability the class regions do not contain the true instances' classes does not exceed e. In practice, however, conformal region classifiers need to be also efficient; i.e., they have to output non-empty and relatively small class regions. In this paper we show that conformal region classification can benefit from instance-transfer learning. Our new approach consists of the basic conformal region classifier with a nonconformity function that implements instance transfer. We propose to learn such a function using a new multi-class Transfer AdaBoost. M1 algorithm. The function and its relation to the conformal region classification are theoretically justified. The experiments showed that our approach is valid for any significance level epsilon is an element of [0, 1] and its efficiency can be improved with instance transfer.

AB - Conformal region classification focuses on developing region classifiers; i.e., classifiers that output regions (sets) of classes for new test instances. 2,13,16 Conformal region classifiers have been proven to be valid for any significance level epsilon is an element of [0, 1] in the sense that the probability the class regions do not contain the true instances' classes does not exceed e. In practice, however, conformal region classifiers need to be also efficient; i.e., they have to output non-empty and relatively small class regions. In this paper we show that conformal region classification can benefit from instance-transfer learning. Our new approach consists of the basic conformal region classifier with a nonconformity function that implements instance transfer. We propose to learn such a function using a new multi-class Transfer AdaBoost. M1 algorithm. The function and its relation to the conformal region classification are theoretically justified. The experiments showed that our approach is valid for any significance level epsilon is an element of [0, 1] and its efficiency can be improved with instance transfer.

U2 - 10.1142/S0218213015600027

DO - 10.1142/S0218213015600027

M3 - Article

VL - 24

JO - International Journal on Artificial Intelligence Tools

JF - International Journal on Artificial Intelligence Tools

SN - 0218-2130

IS - 6

ER -