CaltechAUTHORS
  A Caltech Library Service

A Machine Learning Method to Infer Fundamental Stellar Parameters from Photometric Light Curves

Miller, A. A. and Bloom, J. S. and Richards, J. W. and Lee, Y. S. and Starr, D. L. and Butler, N. R. and Tokarz, S. and Smith, N. and Eisner, J. A. (2014) A Machine Learning Method to Infer Fundamental Stellar Parameters from Photometric Light Curves. . (Submitted) http://resolver.caltech.edu/CaltechAUTHORS:20141117-102958728

[img]
Preview
PDF - Submitted Version
See Usage Policy.

3116Kb

Use this Persistent URL to link to this item: http://resolver.caltech.edu/CaltechAUTHORS:20141117-102958728

Abstract

A fundamental challenge for wide-field imaging surveys is obtaining follow-up spectroscopic observations: there are > 10^9 photometrically cataloged sources, yet modern spectroscopic surveys are limited to ~few x 10^6 targets. As we approach the Large Synoptic Survey Telescope (LSST) era, new algorithmic solutions are required to cope with the data deluge. Here we report the development of a machine-learning framework capable of inferring fundamental stellar parameters (T_(eff), log g, and [Fe/H]) using photometric-brightness variations and color alone. A training set is constructed from a systematic spectroscopic survey of variables with Hectospec/MMT. In sum, the training set includes ~9000 spectra, for which stellar parameters are measured using the SEGUE Stellar Parameters Pipeline (SSPP). We employed the random forest algorithm to perform a non-parametric regression that predicts T_(eff), log g, and [Fe/H] from photometric time-domain observations. Our final, optimized model produces a cross-validated root-mean-square error (RMSE) of 165 K, 0.39 dex, and 0.33 dex for T_(eff), log g, and [Fe/H], respectively. Examining the subset of sources for which the SSPP measurements are most reliable, the RMSE reduces to 125 K, 0.37 dex, and 0.27 dex, respectively, comparable to what is achievable via low-resolution spectroscopy. For variable stars this represents a ≈12-20% improvement in RMSE relative to models trained with single-epoch photometric colors. As an application of our method, we estimate stellar parameters for ~54,000 known variables. We argue that this method may convert photometric time-domain surveys into pseudo-spectrographic engines, enabling the construction of extremely detailed maps of the Milky Way, its structure, and history.


Item Type:Report or Paper (Discussion Paper)
Related URLs:
URLURL TypeDescription
http://arxiv.org/abs/1411.1073arXivDiscussion Paper
Additional Information:DRAFT November 6, 2014. This work has made extensive use of the online data and tools made available by the SDSS collaboration. We are particularly grateful to Z. Ivezic and collaborators at the University of Washington for making their calibrated light curves of Stripe 82 sources publicly available. We thank BD Bue for a fruitful conversation concerning regression bias. We also thank the anonymous referee for several useful comments that have helped to improve this paper. A.A.M. acknowledges support for this work by NASA from a Hubble Fellowship grant: HST-HF-51325.01, awarded by STScI, operated by AURA, Inc., for NASA, under contract NAS 5-26555. J.S.B. acknowledges support from an NSF-CDI grant 0941742. JAE gratefully acknowledges support from an Alfred P. Sloan Research Fellowship. Part of the research was carried out at the Jet Propulsion Laboratory, California Institute of Technology, under a contract with NASA. Observations reported here were obtained at the MMT Observatory, a joint facility of the University of Arizona and the Smithsonian Institution. Facilities: Sloan, MMT (Hectospec)
Funders:
Funding AgencyGrant Number
NASA Hubble FellowshipHST-HF-51325.01
Space Telescope Science InstituteUNSPECIFIED
NASANAS 5-26555
NSF-CDI Grant0941742
Alfred P. Sloan FoundationUNSPECIFIED
NASA/JPL/CaltechUNSPECIFIED
Subject Keywords:methods: data analysis, methods: statistical, stars: general, stars: statistics, stars: variables: general, surveys
Record Number:CaltechAUTHORS:20141117-102958728
Persistent URL:http://resolver.caltech.edu/CaltechAUTHORS:20141117-102958728
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:51836
Collection:CaltechAUTHORS
Deposited By: Ruth Sustaita
Deposited On:17 Nov 2014 21:02
Last Modified:16 Dec 2014 18:08

Repository Staff Only: item control page