Neighbourhood components analysis

Jacob Goldberger; Sam Roweis; Geoff Hinton; Ruslan Salakhutdinov

Neighbourhood components analysis

Jacob Goldberger, Sam Roweis, Geoff Hinton, Ruslan Salakhutdinov

University of Toronto

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

800 Scopus citations

Abstract

In this paper we propose a novel method for learning a Mahalanobis distance measure to be used in the KNN classification algorithm. The algorithm directly maximizes a stochastic variant of the leave-one-out KNN score on the training set. It can also learn a low-dimensional linear embedding of labeled data that can be used for data visualization and fast classification. Unlike other methods, our classification model is non-parametric, making no assumptions about the shape of the class distributions or the boundaries between them. The performance of the method is demonstrated on several data sets, both for metric learning and linear dimensionality reduction.

Original language	English
Title of host publication	Advances in Neural Information Processing Systems 17 - Proceedings of the 2004 Conference, NIPS 2004
Publisher	Neural information processing systems foundation
ISBN (Print)	0262195348, 9780262195348
State	Published - 2005
Externally published	Yes
Event	18th Annual Conference on Neural Information Processing Systems, NIPS 2004 - Vancouver, BC, Canada Duration: 13 Dec 2004 → 16 Dec 2004

Publication series

Name	Advances in Neural Information Processing Systems
ISSN (Print)	1049-5258

Conference

Conference	18th Annual Conference on Neural Information Processing Systems, NIPS 2004
Country/Territory	Canada
City	Vancouver, BC
Period	13/12/04 → 16/12/04

Cite this

@inproceedings{8fe33f96681247229f1c062fc176201e,

title = "Neighbourhood components analysis",

abstract = "In this paper we propose a novel method for learning a Mahalanobis distance measure to be used in the KNN classification algorithm. The algorithm directly maximizes a stochastic variant of the leave-one-out KNN score on the training set. It can also learn a low-dimensional linear embedding of labeled data that can be used for data visualization and fast classification. Unlike other methods, our classification model is non-parametric, making no assumptions about the shape of the class distributions or the boundaries between them. The performance of the method is demonstrated on several data sets, both for metric learning and linear dimensionality reduction.",

author = "Jacob Goldberger and Sam Roweis and Geoff Hinton and Ruslan Salakhutdinov",

year = "2005",

language = "אנגלית",

isbn = "0262195348",

series = "Advances in Neural Information Processing Systems",

publisher = "Neural information processing systems foundation",

booktitle = "Advances in Neural Information Processing Systems 17 - Proceedings of the 2004 Conference, NIPS 2004",

note = "18th Annual Conference on Neural Information Processing Systems, NIPS 2004 ; Conference date: 13-12-2004 Through 16-12-2004",

}

Goldberger, J, Roweis, S, Hinton, G & Salakhutdinov, R 2005, Neighbourhood components analysis. in Advances in Neural Information Processing Systems 17 - Proceedings of the 2004 Conference, NIPS 2004. Advances in Neural Information Processing Systems, Neural information processing systems foundation, 18th Annual Conference on Neural Information Processing Systems, NIPS 2004, Vancouver, BC, Canada, 13/12/04.

Neighbourhood components analysis. / Goldberger, Jacob; Roweis, Sam; Hinton, Geoff et al.
Advances in Neural Information Processing Systems 17 - Proceedings of the 2004 Conference, NIPS 2004. Neural information processing systems foundation, 2005. (Advances in Neural Information Processing Systems).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Neighbourhood components analysis

AU - Goldberger, Jacob

AU - Roweis, Sam

AU - Hinton, Geoff

AU - Salakhutdinov, Ruslan

PY - 2005

Y1 - 2005

N2 - In this paper we propose a novel method for learning a Mahalanobis distance measure to be used in the KNN classification algorithm. The algorithm directly maximizes a stochastic variant of the leave-one-out KNN score on the training set. It can also learn a low-dimensional linear embedding of labeled data that can be used for data visualization and fast classification. Unlike other methods, our classification model is non-parametric, making no assumptions about the shape of the class distributions or the boundaries between them. The performance of the method is demonstrated on several data sets, both for metric learning and linear dimensionality reduction.

AB - In this paper we propose a novel method for learning a Mahalanobis distance measure to be used in the KNN classification algorithm. The algorithm directly maximizes a stochastic variant of the leave-one-out KNN score on the training set. It can also learn a low-dimensional linear embedding of labeled data that can be used for data visualization and fast classification. Unlike other methods, our classification model is non-parametric, making no assumptions about the shape of the class distributions or the boundaries between them. The performance of the method is demonstrated on several data sets, both for metric learning and linear dimensionality reduction.

UR - http://www.scopus.com/inward/record.url?scp=84898993653&partnerID=8YFLogxK

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:84898993653

SN - 0262195348

SN - 9780262195348

T3 - Advances in Neural Information Processing Systems

BT - Advances in Neural Information Processing Systems 17 - Proceedings of the 2004 Conference, NIPS 2004

PB - Neural information processing systems foundation

T2 - 18th Annual Conference on Neural Information Processing Systems, NIPS 2004

Y2 - 13 December 2004 through 16 December 2004

ER -

Neighbourhood components analysis

Abstract

Publication series

Conference

Other files and links

Fingerprint

Cite this