Nearest Class Vector Classification for Large-Scale Learning Problems

Alexandros Iosifidis; Anastasios Tefas; Ioannis Pitas

doi:10.1109/Trustcom.2015.556

Nearest Class Vector Classification for Large-Scale Learning Problems

Alexandros Iosifidis, Anastasios Tefas, Ioannis Pitas

Department of Electrical & Electronic Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

319 Downloads (Pure)

Abstract

In this paper, we describe a method for combined metric learning and classification, that is based on logistic discrimination for the determination of a low-dimensional feature space of increased discrimination power. An iterating optimization process is applied to this end, where the probability of correct classification rate is increased at each optimization step. Extensions of the method that allow richer class representations and non-linear feature space determination and classification are also described. The described optimization schemes are solved by following (stochastic or mini-batch) gradient descent optimization, which is well suited for large-scale learning problems.

Original language	English
Title of host publication	2015 IEEE Trustcom/BigDataSE/ISPA
Subtitle of host publication	Proceedings of a meeting held 20-22 August 2015, Helsinki, Finland
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Pages	11-16
Number of pages	6
Volume	2
ISBN (Electronic)	9781467379526
ISBN (Print)	9781467379533
DOIs	https://doi.org/10.1109/Trustcom.2015.556
Publication status	Published - Jan 2016
Event	IEEE International Conference on Big Data Science and Engineering (BigDataSE) - Helsinki, Finland Duration: 20 Aug 2015 → 22 Aug 2015

Conference

Conference	IEEE International Conference on Big Data Science and Engineering (BigDataSE)
Country/Territory	Finland
City	Helsinki
Period	20/08/15 → 22/08/15

Keywords

Nearest Class Vector classification
Logistic Discrimination

Access to Document

10.1109/Trustcom.2015.556

Ioannis Pitas - Nearest class vector classification for large-scale learning problems - 2015Accepted author manuscript, 2.02 MB

Handle.net

Persistent link

Cite this

@inproceedings{54931831adb8474fa7ac11467725d80f,

title = "Nearest Class Vector Classification for Large-Scale Learning Problems",

abstract = "In this paper, we describe a method for combined metric learning and classification, that is based on logistic discrimination for the determination of a low-dimensional feature space of increased discrimination power. An iterating optimization process is applied to this end, where the probability of correct classification rate is increased at each optimization step. Extensions of the method that allow richer class representations and non-linear feature space determination and classification are also described. The described optimization schemes are solved by following (stochastic or mini-batch) gradient descent optimization, which is well suited for large-scale learning problems.",

keywords = "Nearest Class Vector classification, Logistic Discrimination",

author = "Alexandros Iosifidis and Anastasios Tefas and Ioannis Pitas",

year = "2016",

month = jan,

doi = "10.1109/Trustcom.2015.556",

language = "English",

isbn = "9781467379533",

volume = "2",

pages = "11--16",

booktitle = "2015 IEEE Trustcom/BigDataSE/ISPA",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

address = "United States",

note = "IEEE International Conference on Big Data Science and Engineering (BigDataSE) ; Conference date: 20-08-2015 Through 22-08-2015",

}

Iosifidis, A, Tefas, A & Pitas, I 2016, Nearest Class Vector Classification for Large-Scale Learning Problems. in 2015 IEEE Trustcom/BigDataSE/ISPA: Proceedings of a meeting held 20-22 August 2015, Helsinki, Finland. vol. 2, Institute of Electrical and Electronics Engineers (IEEE), pp. 11-16, IEEE International Conference on Big Data Science and Engineering (BigDataSE), Helsinki, Finland, 20/08/15. https://doi.org/10.1109/Trustcom.2015.556

Nearest Class Vector Classification for Large-Scale Learning Problems. / Iosifidis, Alexandros; Tefas, Anastasios; Pitas, Ioannis.
2015 IEEE Trustcom/BigDataSE/ISPA: Proceedings of a meeting held 20-22 August 2015, Helsinki, Finland. Vol. 2 Institute of Electrical and Electronics Engineers (IEEE), 2016. p. 11-16.

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

TY - GEN

T1 - Nearest Class Vector Classification for Large-Scale Learning Problems

AU - Iosifidis, Alexandros

AU - Tefas, Anastasios

AU - Pitas, Ioannis

PY - 2016/1

Y1 - 2016/1

N2 - In this paper, we describe a method for combined metric learning and classification, that is based on logistic discrimination for the determination of a low-dimensional feature space of increased discrimination power. An iterating optimization process is applied to this end, where the probability of correct classification rate is increased at each optimization step. Extensions of the method that allow richer class representations and non-linear feature space determination and classification are also described. The described optimization schemes are solved by following (stochastic or mini-batch) gradient descent optimization, which is well suited for large-scale learning problems.

AB - In this paper, we describe a method for combined metric learning and classification, that is based on logistic discrimination for the determination of a low-dimensional feature space of increased discrimination power. An iterating optimization process is applied to this end, where the probability of correct classification rate is increased at each optimization step. Extensions of the method that allow richer class representations and non-linear feature space determination and classification are also described. The described optimization schemes are solved by following (stochastic or mini-batch) gradient descent optimization, which is well suited for large-scale learning problems.

KW - Nearest Class Vector classification

KW - Logistic Discrimination

U2 - 10.1109/Trustcom.2015.556

DO - 10.1109/Trustcom.2015.556

M3 - Conference Contribution (Conference Proceeding)

SN - 9781467379533

VL - 2

SP - 11

EP - 16

BT - 2015 IEEE Trustcom/BigDataSE/ISPA

PB - Institute of Electrical and Electronics Engineers (IEEE)

T2 - IEEE International Conference on Big Data Science and Engineering (BigDataSE)

Y2 - 20 August 2015 through 22 August 2015

ER -

Nearest Class Vector Classification for Large-Scale Learning Problems

Abstract

Conference

Keywords

Access to Document

Handle.net

Fingerprint

Cite this