Deep auto-encoder in model reduction of lage-scale spatiotemporal dynamics

Mingliang WANG; Han-Xiong LI; Wenjing SHEN

doi:10.1109/IJCNN.2016.7727605

Deep auto-encoder in model reduction of lage-scale spatiotemporal dynamics

Mingliang WANG, Han-Xiong LI, Wenjing SHEN

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

10 Citations (Scopus)

Abstract

This paper presents a deep auto-encoder based model reduction method for large scale spatiotemporal process. This method includes three phases in order to find the near-optimal parameters of the reduced order model. The sequence of the phases is allocated according to the idea of greedy training which approximately minimizes the modeling error. This method also avoids including the spatial dimensionality into the model which enables it to handle large-scale model reduction. Two case studies are carried out to demonstrate the effectiveness of the method.

Original language	English
Title of host publication	2016 International Joint Conference on Neural Networks (IJCNN)
Publisher	IEEE
Pages	3180-3186
ISBN (Electronic)	978-1-5090-0620-5
ISBN (Print)	9781509006199
DOIs	https://doi.org/10.1109/IJCNN.2016.7727605
Publication status	Published - Jul 2016
Event	2016 International Joint Conference on Neural Networks (IJCNN 2016) - Vancouver Convention Centre , Vancouver, Canada Duration: 24 Jul 2016 → 29 Jul 2016 http://www.wcci2016.org/

Publication series

Name
ISSN (Electronic)	2161-4407

Conference

Conference	2016 International Joint Conference on Neural Networks (IJCNN 2016)
Abbreviated title	IJCNN 2016
Place	Canada
City	Vancouver
Period	24/07/16 → 29/07/16
Internet address	http://www.wcci2016.org/

Research Keywords

Deep auto-encoder
Model reduction
Restricted Boltzmann Machine

Access Output

10.1109/IJCNN.2016.7727605

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

@inproceedings{346c062517854ac3a2ffd0825c2c9534,

title = "Deep auto-encoder in model reduction of lage-scale spatiotemporal dynamics",

abstract = "This paper presents a deep auto-encoder based model reduction method for large scale spatiotemporal process. This method includes three phases in order to find the near-optimal parameters of the reduced order model. The sequence of the phases is allocated according to the idea of greedy training which approximately minimizes the modeling error. This method also avoids including the spatial dimensionality into the model which enables it to handle large-scale model reduction. Two case studies are carried out to demonstrate the effectiveness of the method.",

keywords = "Deep auto-encoder, Model reduction, Restricted Boltzmann Machine",

author = "Mingliang WANG and Han-Xiong LI and Wenjing SHEN",

year = "2016",

month = jul,

doi = "10.1109/IJCNN.2016.7727605",

language = "English",

isbn = "9781509006199",

publisher = "IEEE",

pages = "3180--3186",

booktitle = "2016 International Joint Conference on Neural Networks (IJCNN)",

address = "United States",

note = "2016 International Joint Conference on Neural Networks (IJCNN 2016), IJCNN 2016 ; Conference date: 24-07-2016 Through 29-07-2016",

url = "http://www.wcci2016.org/",

}

Deep auto-encoder in model reduction of lage-scale spatiotemporal dynamics. / WANG, Mingliang; LI, Han-Xiong ; SHEN, Wenjing.
2016 International Joint Conference on Neural Networks (IJCNN). IEEE, 2016. p. 3180-3186 7727605.

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - Deep auto-encoder in model reduction of lage-scale spatiotemporal dynamics

AU - WANG, Mingliang

AU - LI, Han-Xiong

AU - SHEN, Wenjing

PY - 2016/7

Y1 - 2016/7

N2 - This paper presents a deep auto-encoder based model reduction method for large scale spatiotemporal process. This method includes three phases in order to find the near-optimal parameters of the reduced order model. The sequence of the phases is allocated according to the idea of greedy training which approximately minimizes the modeling error. This method also avoids including the spatial dimensionality into the model which enables it to handle large-scale model reduction. Two case studies are carried out to demonstrate the effectiveness of the method.

AB - This paper presents a deep auto-encoder based model reduction method for large scale spatiotemporal process. This method includes three phases in order to find the near-optimal parameters of the reduced order model. The sequence of the phases is allocated according to the idea of greedy training which approximately minimizes the modeling error. This method also avoids including the spatial dimensionality into the model which enables it to handle large-scale model reduction. Two case studies are carried out to demonstrate the effectiveness of the method.

KW - Deep auto-encoder

KW - Model reduction

KW - Restricted Boltzmann Machine

UR - http://www.scopus.com/inward/record.url?scp=85007275587&partnerID=8YFLogxK

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85007275587&origin=recordpage

U2 - 10.1109/IJCNN.2016.7727605

DO - 10.1109/IJCNN.2016.7727605

M3 - RGC 32 - Refereed conference paper (with host publication)

SN - 9781509006199

SP - 3180

EP - 3186

BT - 2016 International Joint Conference on Neural Networks (IJCNN)

PB - IEEE

T2 - 2016 International Joint Conference on Neural Networks (IJCNN 2016)

Y2 - 24 July 2016 through 29 July 2016

ER -

Deep auto-encoder in model reduction of lage-scale spatiotemporal dynamics

Abstract

Publication series

Conference

Research Keywords

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this