Cloud clustering over January 2003 via scalable rotationally invariant autoencoder

Takuya Kurihana; Elisabeth Moyer; Rebecca Willett; Davis Gilton; Ian Foster

doi:10.1109/eScience51609.2021.00047

Cloud clustering over January 2003 via scalable rotationally invariant autoencoder

Takuya Kurihana, Elisabeth Moyer, Rebecca Willett, Davis Gilton, Ian Foster

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

Abstract

Unsupervised fashion of cloud analysis has the significant possibility of exploring massive quantities of satellite cloud imagery to discover unknown cloud patterns that can be relevant to climate change research, free from the assumption of artificial cloud categories. We describe a further development of rotation-invariant cloud clustering (RICC) that leverages unsupervised deep learning autoencoder and clustering to be scaled for larger cloud datasets. Results suggest that our rotation-invariant autoencoder shows high scalability conditioned on the size of GPUs, and the clusters generated from RICC on the month-long dataset capture unique spatial patterns with distinct cloud physical properties.

Original language	English
Title of host publication	Proceedings - IEEE 17th International Conference on eScience, eScience 2021
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	253-254
Number of pages	2
ISBN (Electronic)	9781665403610
DOIs	https://doi.org/10.1109/eScience51609.2021.00047
State	Published - Sep 2021
Externally published	Yes
Event	17th IEEE International Conference on eScience, eScience 2021 - Virtual, Online, Austria Duration: Sep 20 2021 → Sep 23 2021

Publication series

Name	Proceedings - IEEE 17th International Conference on eScience, eScience 2021

Conference

Conference	17th IEEE International Conference on eScience, eScience 2021
Country/Territory	Austria
City	Virtual, Online
Period	09/20/21 → 09/23/21

Keywords

Autoencoder
Cloud classification
Clustering

Access to Document

10.1109/eScience51609.2021.00047

Cite this

Kurihana, T., Moyer, E., Willett, R., Gilton, D., & Foster, I. (2021). Cloud clustering over January 2003 via scalable rotationally invariant autoencoder. In Proceedings - IEEE 17th International Conference on eScience, eScience 2021 (pp. 253-254). (Proceedings - IEEE 17th International Conference on eScience, eScience 2021). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/eScience51609.2021.00047

Kurihana, Takuya ; Moyer, Elisabeth ; Willett, Rebecca et al. / Cloud clustering over January 2003 via scalable rotationally invariant autoencoder. Proceedings - IEEE 17th International Conference on eScience, eScience 2021. Institute of Electrical and Electronics Engineers Inc., 2021. pp. 253-254 (Proceedings - IEEE 17th International Conference on eScience, eScience 2021).

@inproceedings{ac4364d2d275499b9d020a779e8a18ed,

title = "Cloud clustering over January 2003 via scalable rotationally invariant autoencoder",

abstract = "Unsupervised fashion of cloud analysis has the significant possibility of exploring massive quantities of satellite cloud imagery to discover unknown cloud patterns that can be relevant to climate change research, free from the assumption of artificial cloud categories. We describe a further development of rotation-invariant cloud clustering (RICC) that leverages unsupervised deep learning autoencoder and clustering to be scaled for larger cloud datasets. Results suggest that our rotation-invariant autoencoder shows high scalability conditioned on the size of GPUs, and the clusters generated from RICC on the month-long dataset capture unique spatial patterns with distinct cloud physical properties.",

keywords = "Autoencoder, Cloud classification, Clustering",

author = "Takuya Kurihana and Elisabeth Moyer and Rebecca Willett and Davis Gilton and Ian Foster",

note = "Publisher Copyright: {\textcopyright} 2021 IEEE.; 17th IEEE International Conference on eScience, eScience 2021 ; Conference date: 20-09-2021 Through 23-09-2021",

year = "2021",

month = sep,

doi = "10.1109/eScience51609.2021.00047",

language = "English",

series = "Proceedings - IEEE 17th International Conference on eScience, eScience 2021",

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

pages = "253--254",

booktitle = "Proceedings - IEEE 17th International Conference on eScience, eScience 2021",

}

Kurihana, T, Moyer, E, Willett, R, Gilton, D & Foster, I 2021, Cloud clustering over January 2003 via scalable rotationally invariant autoencoder. in Proceedings - IEEE 17th International Conference on eScience, eScience 2021. Proceedings - IEEE 17th International Conference on eScience, eScience 2021, Institute of Electrical and Electronics Engineers Inc., pp. 253-254, 17th IEEE International Conference on eScience, eScience 2021, Virtual, Online, Austria, 09/20/21. https://doi.org/10.1109/eScience51609.2021.00047

Cloud clustering over January 2003 via scalable rotationally invariant autoencoder. / Kurihana, Takuya; Moyer, Elisabeth; Willett, Rebecca et al.
Proceedings - IEEE 17th International Conference on eScience, eScience 2021. Institute of Electrical and Electronics Engineers Inc., 2021. p. 253-254 (Proceedings - IEEE 17th International Conference on eScience, eScience 2021).

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

TY - GEN

T1 - Cloud clustering over January 2003 via scalable rotationally invariant autoencoder

AU - Kurihana, Takuya

AU - Moyer, Elisabeth

AU - Willett, Rebecca

AU - Gilton, Davis

AU - Foster, Ian

PY - 2021/9

Y1 - 2021/9

N2 - Unsupervised fashion of cloud analysis has the significant possibility of exploring massive quantities of satellite cloud imagery to discover unknown cloud patterns that can be relevant to climate change research, free from the assumption of artificial cloud categories. We describe a further development of rotation-invariant cloud clustering (RICC) that leverages unsupervised deep learning autoencoder and clustering to be scaled for larger cloud datasets. Results suggest that our rotation-invariant autoencoder shows high scalability conditioned on the size of GPUs, and the clusters generated from RICC on the month-long dataset capture unique spatial patterns with distinct cloud physical properties.

AB - Unsupervised fashion of cloud analysis has the significant possibility of exploring massive quantities of satellite cloud imagery to discover unknown cloud patterns that can be relevant to climate change research, free from the assumption of artificial cloud categories. We describe a further development of rotation-invariant cloud clustering (RICC) that leverages unsupervised deep learning autoencoder and clustering to be scaled for larger cloud datasets. Results suggest that our rotation-invariant autoencoder shows high scalability conditioned on the size of GPUs, and the clusters generated from RICC on the month-long dataset capture unique spatial patterns with distinct cloud physical properties.

KW - Autoencoder

KW - Cloud classification

KW - Clustering

UR - http://www.scopus.com/inward/record.url?scp=85119088117&partnerID=8YFLogxK

U2 - 10.1109/eScience51609.2021.00047

DO - 10.1109/eScience51609.2021.00047

M3 - Conference contribution

AN - SCOPUS:85119088117

T3 - Proceedings - IEEE 17th International Conference on eScience, eScience 2021

SP - 253

EP - 254

BT - Proceedings - IEEE 17th International Conference on eScience, eScience 2021

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 17th IEEE International Conference on eScience, eScience 2021

Y2 - 20 September 2021 through 23 September 2021

ER -

Kurihana T, Moyer E, Willett R, Gilton D, Foster I. Cloud clustering over January 2003 via scalable rotationally invariant autoencoder. In Proceedings - IEEE 17th International Conference on eScience, eScience 2021. Institute of Electrical and Electronics Engineers Inc. 2021. p. 253-254. (Proceedings - IEEE 17th International Conference on eScience, eScience 2021). doi: 10.1109/eScience51609.2021.00047

Cloud clustering over January 2003 via scalable rotationally invariant autoencoder

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this