Machine Learning for the Zwicky Transient Facility

The Zwicky Transient Facility is a large optical survey in multiple filters producing hundreds of thousands of transient alerts per night. We describe here various machine learning (ML) implementations and plans to make the maximal use of the large data set by taking advantage of the temporal nature...

Full description

Saved in:
Bibliographic Details
Published in:Publications of the Astronomical Society of the Pacific Vol. 131; no. 997
Main Authors: Mahabal, Ashish, Rebbapragada, Umaa, Walters, Richard, Masci, Frank J., Blagorodnova, Nadejda, Roestel, Jan van, Ye, Quan-Zhi, Biswas, Rahul, Burdge, Kevin, Chang, Chan-Kao, Duev, Dmitry A., Golkhou, V. Zach, Miller, Adam A., Nordin, Jakob, Ward, Charlotte, Adams, Scott, Bellm, Eric C., Branton, Doug, Bue, Brian, Cannella, Chris, Connolly, Andrew, Dekany, Richard, Feindt, Ulrich, Hung, Tiara, Fortson, Lucy, Frederick, Sara, Fremling, C., Gezari, Suvi, Graham, Matthew, Groom, Steven, Kasliwal, Mansi M., Kulkarni, Shrinivas, Kupfer, Thomas, Lin, Hsing Wen, Lintott, Chris, Lunnan, Ragnhild, Parejko, John, Prince, Thomas A., Riddle, Reed, Rusholme, Ben, Saunders, Nicholas, Sedaghat, Nima, Shupe, David L., Singer, Leo P., Soumagnac, Maayane T., Szkody, Paula, Tachibana, Yutaro, Tirumala, Kushal, Velzen, Sjoert van, Wright, Darryl
Format: Journal Article
Language:English
Published: The Astronomical Society of the Pacific 01.03.2019
Subjects:
Machine Learning
Sky Surveys
Time Domain
ISSN:0004-6280, 1538-3873
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The Zwicky Transient Facility is a large optical survey in multiple filters producing hundreds of thousands of transient alerts per night. We describe here various machine learning (ML) implementations and plans to make the maximal use of the large data set by taking advantage of the temporal nature of the data, and further combining it with other data sets. We start with the initial steps of separating bogus candidates from real ones, separating stars and galaxies, and go on to the classification of real objects into various classes. Besides the usual methods (e.g., based on features extracted from light curves) we also describe early plans for alternate methods including the use of domain adaptation, and deep learning. In a similar fashion we describe efforts to detect fast moving asteroids. We also describe the use of the Zooniverse platform for helping with classifications through the creation of training samples, and active learning. Finally we mention the synergistic aspects of ZTF and LSST from the ML perspective.
Bibliography:PASP-100652.R2
ISSN:0004-6280
1538-3873
DOI:10.1088/1538-3873/aaf3fa