Abstract

In this chapter, we describe how the p-values derived from the conformal predictions framework can be used for active learning; that is, to select the informative examples from a data collection that can be used to train a classifier for best performance. We show the connection of this approach to information-theoretic methods, as well as show how the methodology can be generalized to multiple classifier models and information fusion settings.

Original languageEnglish (US)
Title of host publicationConformal Prediction for Reliable Machine Learning: Theory, Adaptations and Applications
PublisherElsevier Inc.
Pages49-70
Number of pages22
ISBN (Print)9780123985378
DOIs
StatePublished - Apr 2014

Fingerprint

Classifiers
Information fusion
Problem-Based Learning

Keywords

  • Active Learning
  • Face Recognition
  • Image Classification
  • Multicriteria Active Learning
  • Online Active Learning
  • Query by Transduction

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Balasubramanian, V. N., Chakraborty, S., Ho, S. S., Wechsler, H., & Panchanathan, S. (2014). Active Learning. In Conformal Prediction for Reliable Machine Learning: Theory, Adaptations and Applications (pp. 49-70). Elsevier Inc.. https://doi.org/10.1016/B978-0-12-398537-8.00003-1

Active Learning. / Balasubramanian, Vineeth N.; Chakraborty, Shayok; Ho, Shen Shyang; Wechsler, Harry; Panchanathan, Sethuraman.

Conformal Prediction for Reliable Machine Learning: Theory, Adaptations and Applications. Elsevier Inc., 2014. p. 49-70.

Research output: Chapter in Book/Report/Conference proceedingChapter

Balasubramanian, VN, Chakraborty, S, Ho, SS, Wechsler, H & Panchanathan, S 2014, Active Learning. in Conformal Prediction for Reliable Machine Learning: Theory, Adaptations and Applications. Elsevier Inc., pp. 49-70. https://doi.org/10.1016/B978-0-12-398537-8.00003-1
Balasubramanian VN, Chakraborty S, Ho SS, Wechsler H, Panchanathan S. Active Learning. In Conformal Prediction for Reliable Machine Learning: Theory, Adaptations and Applications. Elsevier Inc. 2014. p. 49-70 https://doi.org/10.1016/B978-0-12-398537-8.00003-1
Balasubramanian, Vineeth N. ; Chakraborty, Shayok ; Ho, Shen Shyang ; Wechsler, Harry ; Panchanathan, Sethuraman. / Active Learning. Conformal Prediction for Reliable Machine Learning: Theory, Adaptations and Applications. Elsevier Inc., 2014. pp. 49-70
@inbook{d1fe3822081a4a268cc8025d18f14474,
title = "Active Learning",
abstract = "In this chapter, we describe how the p-values derived from the conformal predictions framework can be used for active learning; that is, to select the informative examples from a data collection that can be used to train a classifier for best performance. We show the connection of this approach to information-theoretic methods, as well as show how the methodology can be generalized to multiple classifier models and information fusion settings.",
keywords = "Active Learning, Face Recognition, Image Classification, Multicriteria Active Learning, Online Active Learning, Query by Transduction",
author = "Balasubramanian, {Vineeth N.} and Shayok Chakraborty and Ho, {Shen Shyang} and Harry Wechsler and Sethuraman Panchanathan",
year = "2014",
month = "4",
doi = "10.1016/B978-0-12-398537-8.00003-1",
language = "English (US)",
isbn = "9780123985378",
pages = "49--70",
booktitle = "Conformal Prediction for Reliable Machine Learning: Theory, Adaptations and Applications",
publisher = "Elsevier Inc.",
address = "United States",

}

TY - CHAP

T1 - Active Learning

AU - Balasubramanian, Vineeth N.

AU - Chakraborty, Shayok

AU - Ho, Shen Shyang

AU - Wechsler, Harry

AU - Panchanathan, Sethuraman

PY - 2014/4

Y1 - 2014/4

N2 - In this chapter, we describe how the p-values derived from the conformal predictions framework can be used for active learning; that is, to select the informative examples from a data collection that can be used to train a classifier for best performance. We show the connection of this approach to information-theoretic methods, as well as show how the methodology can be generalized to multiple classifier models and information fusion settings.

AB - In this chapter, we describe how the p-values derived from the conformal predictions framework can be used for active learning; that is, to select the informative examples from a data collection that can be used to train a classifier for best performance. We show the connection of this approach to information-theoretic methods, as well as show how the methodology can be generalized to multiple classifier models and information fusion settings.

KW - Active Learning

KW - Face Recognition

KW - Image Classification

KW - Multicriteria Active Learning

KW - Online Active Learning

KW - Query by Transduction

UR - http://www.scopus.com/inward/record.url?scp=84902228612&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84902228612&partnerID=8YFLogxK

U2 - 10.1016/B978-0-12-398537-8.00003-1

DO - 10.1016/B978-0-12-398537-8.00003-1

M3 - Chapter

SN - 9780123985378

SP - 49

EP - 70

BT - Conformal Prediction for Reliable Machine Learning: Theory, Adaptations and Applications

PB - Elsevier Inc.

ER -