Deep-learnt classification of light curves
Abstract
Astronomy light curves are sparse, gappy, and heteroscedastic. As a result standard time series methods regularly used for financial and similar datasets are of little help and astronomers are usually left to their own instruments and techniques to classify light curves. A common approach is to derive statistical features from the time series and to use machine learning methods, generally supervised, to separate objects into a few of the standard classes. In this work, we transform the time series to two-dimensional light curve representations in order to classify them using modern deep learning techniques. In particular, we show that convolutional neural networks based classifiers work well for broad characterization and classification. We use labeled datasets of periodic variables from CRTS survey and show how this opens doors for a quick classification of diverse classes with several possible exciting extensions.
Additional Information
© 2017 IEEE. This work, and CRTS survey, was supported in part by the NSF grants AST-0909182, AST-1313422, AST-1413600, and AST-1518308, and by the Ajax Foundation. KS thanks IIT Gandhinagar and the Caltech SURF program.Attached Files
Submitted - 1709.06257.pdf
Files
Name | Size | Download all |
---|---|---|
md5:ba54ab2511d1e0df42a6c3c46a9833b7
|
993.0 kB | Preview Download |
Additional details
- Eprint ID
- 84742
- DOI
- 10.1109/SSCI.2017.8280984
- Resolver ID
- CaltechAUTHORS:20180208-145104828
- NSF
- AST-0909182
- NSF
- AST-1313422
- NSF
- AST-1413600
- NSF
- AST-1518308
- Ajax Foundation
- Caltech Summer Undergraduate Research Fellowship (SURF)
- Created
-
2018-02-09Created from EPrint's datestamp field
- Updated
-
2021-11-15Created from EPrint's last_modified field
- Caltech groups
- Astronomy Department