Extending out-of-sample manifold learning via meta-modelling techniques

Gulsen Taskin; Melba Crawford

doi:10.1109/IGARSS.2017.8127015

Extending out-of-sample manifold learning via meta-modelling techniques

Gulsen Taskin^*, Melba Crawford

^*Corresponding author for this work

Disaster Management Institute

Purdue University

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

4 Citations (Scopus)

Abstract

Unsupervised manifold learning has become accepted as an important tool for reducing dimensionality of a data set by finding its meaningful low dimensional representation lying on an unknown nonlinear subspace. Most manifold learning methods only embed an existing data set, but do not provide an explicit mapping function for novel out-of-sample data, thereby potentially resulting in an ineffective tool for classification purposes. To address this issue, out-of-sample extension methods have been introduced to generalize an existing embedding to new samples. In this work, a meta-modelling method called High Dimensional Model Representation (HDMR) is firstly implemented as a nonlinear multivariate regression for the out-of-sample problem for non-parametric unsupervised manifold learning algorithms. Several experiments show that the proposed method outperforms several state-of-the-art out-of-sample extension methods in terms of generalization to new samples for classification experiments on two remote sensing hyperspectral data sets.

Original language	English
Title of host publication	2017 IEEE International Geoscience and Remote Sensing Symposium
Subtitle of host publication	International Cooperation for Global Awareness, IGARSS 2017 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	562-565
Number of pages	4
ISBN (Electronic)	9781509049516
DOIs	https://doi.org/10.1109/IGARSS.2017.8127015
Publication status	Published - 1 Dec 2017
Event	37th Annual IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2017 - Fort Worth, United States Duration: 23 Jul 2017 → 28 Jul 2017

Publication series

Name	International Geoscience and Remote Sensing Symposium (IGARSS)
Volume	2017-July

Conference

Conference	37th Annual IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2017
Country/Territory	United States
City	Fort Worth
Period	23/07/17 → 28/07/17

Bibliographical note

Publisher Copyright:
© 2017 IEEE.

Funding

∗The first author would like to thank to The Scientific and Technological Research Council of Turkey (TUBITAK) 2219 Program for its financial support. This work was performed at School of Civil Engineering in Purdue University.

Funders	Funder number
Scientific and Technological Research Council of Turkey
TUBITAK

Keywords

manifold learning
multivariate regression and classification
Nonlinear dimensionality reduction
out-of-sample extension

Access to Document

10.1109/IGARSS.2017.8127015

Cite this

Taskin, G., & Crawford, M. (2017). Extending out-of-sample manifold learning via meta-modelling techniques. In 2017 IEEE International Geoscience and Remote Sensing Symposium: International Cooperation for Global Awareness, IGARSS 2017 - Proceedings (pp. 562-565). Article 8127015 (International Geoscience and Remote Sensing Symposium (IGARSS); Vol. 2017-July). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IGARSS.2017.8127015

Taskin, Gulsen ; Crawford, Melba. / Extending out-of-sample manifold learning via meta-modelling techniques. 2017 IEEE International Geoscience and Remote Sensing Symposium: International Cooperation for Global Awareness, IGARSS 2017 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 562-565 (International Geoscience and Remote Sensing Symposium (IGARSS)).

@inproceedings{20cd3334434d4ccda57a9c64980cf011,

title = "Extending out-of-sample manifold learning via meta-modelling techniques",

abstract = "Unsupervised manifold learning has become accepted as an important tool for reducing dimensionality of a data set by finding its meaningful low dimensional representation lying on an unknown nonlinear subspace. Most manifold learning methods only embed an existing data set, but do not provide an explicit mapping function for novel out-of-sample data, thereby potentially resulting in an ineffective tool for classification purposes. To address this issue, out-of-sample extension methods have been introduced to generalize an existing embedding to new samples. In this work, a meta-modelling method called High Dimensional Model Representation (HDMR) is firstly implemented as a nonlinear multivariate regression for the out-of-sample problem for non-parametric unsupervised manifold learning algorithms. Several experiments show that the proposed method outperforms several state-of-the-art out-of-sample extension methods in terms of generalization to new samples for classification experiments on two remote sensing hyperspectral data sets.",

keywords = "manifold learning, multivariate regression and classification, Nonlinear dimensionality reduction, out-of-sample extension",

author = "Gulsen Taskin and Melba Crawford",

note = "Publisher Copyright: {\textcopyright} 2017 IEEE.; 37th Annual IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2017 ; Conference date: 23-07-2017 Through 28-07-2017",

year = "2017",

month = dec,

day = "1",

doi = "10.1109/IGARSS.2017.8127015",

language = "English",

series = "International Geoscience and Remote Sensing Symposium (IGARSS)",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "562--565",

booktitle = "2017 IEEE International Geoscience and Remote Sensing Symposium",

address = "United States",

}

Taskin, G & Crawford, M 2017, Extending out-of-sample manifold learning via meta-modelling techniques. in 2017 IEEE International Geoscience and Remote Sensing Symposium: International Cooperation for Global Awareness, IGARSS 2017 - Proceedings., 8127015, International Geoscience and Remote Sensing Symposium (IGARSS), vol. 2017-July, Institute of Electrical and Electronics Engineers Inc., pp. 562-565, 37th Annual IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2017, Fort Worth, United States, 23/07/17. https://doi.org/10.1109/IGARSS.2017.8127015

Extending out-of-sample manifold learning via meta-modelling techniques. / Taskin, Gulsen; Crawford, Melba.
2017 IEEE International Geoscience and Remote Sensing Symposium: International Cooperation for Global Awareness, IGARSS 2017 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2017. p. 562-565 8127015 (International Geoscience and Remote Sensing Symposium (IGARSS); Vol. 2017-July).

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

TY - GEN

T1 - Extending out-of-sample manifold learning via meta-modelling techniques

AU - Taskin, Gulsen

AU - Crawford, Melba

PY - 2017/12/1

Y1 - 2017/12/1

N2 - Unsupervised manifold learning has become accepted as an important tool for reducing dimensionality of a data set by finding its meaningful low dimensional representation lying on an unknown nonlinear subspace. Most manifold learning methods only embed an existing data set, but do not provide an explicit mapping function for novel out-of-sample data, thereby potentially resulting in an ineffective tool for classification purposes. To address this issue, out-of-sample extension methods have been introduced to generalize an existing embedding to new samples. In this work, a meta-modelling method called High Dimensional Model Representation (HDMR) is firstly implemented as a nonlinear multivariate regression for the out-of-sample problem for non-parametric unsupervised manifold learning algorithms. Several experiments show that the proposed method outperforms several state-of-the-art out-of-sample extension methods in terms of generalization to new samples for classification experiments on two remote sensing hyperspectral data sets.

AB - Unsupervised manifold learning has become accepted as an important tool for reducing dimensionality of a data set by finding its meaningful low dimensional representation lying on an unknown nonlinear subspace. Most manifold learning methods only embed an existing data set, but do not provide an explicit mapping function for novel out-of-sample data, thereby potentially resulting in an ineffective tool for classification purposes. To address this issue, out-of-sample extension methods have been introduced to generalize an existing embedding to new samples. In this work, a meta-modelling method called High Dimensional Model Representation (HDMR) is firstly implemented as a nonlinear multivariate regression for the out-of-sample problem for non-parametric unsupervised manifold learning algorithms. Several experiments show that the proposed method outperforms several state-of-the-art out-of-sample extension methods in terms of generalization to new samples for classification experiments on two remote sensing hyperspectral data sets.

KW - manifold learning

KW - multivariate regression and classification

KW - Nonlinear dimensionality reduction

KW - out-of-sample extension

UR - http://www.scopus.com/inward/record.url?scp=85041820721&partnerID=8YFLogxK

U2 - 10.1109/IGARSS.2017.8127015

DO - 10.1109/IGARSS.2017.8127015

M3 - Conference contribution

AN - SCOPUS:85041820721

T3 - International Geoscience and Remote Sensing Symposium (IGARSS)

SP - 562

EP - 565

BT - 2017 IEEE International Geoscience and Remote Sensing Symposium

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 37th Annual IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2017

Y2 - 23 July 2017 through 28 July 2017

ER -

Taskin G, Crawford M. Extending out-of-sample manifold learning via meta-modelling techniques. In 2017 IEEE International Geoscience and Remote Sensing Symposium: International Cooperation for Global Awareness, IGARSS 2017 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2017. p. 562-565. 8127015. (International Geoscience and Remote Sensing Symposium (IGARSS)). doi: 10.1109/IGARSS.2017.8127015

Extending out-of-sample manifold learning via meta-modelling techniques

Abstract

Publication series

Conference

Bibliographical note

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this