The Galah Survey: Classification and diagnostics with t-SNE reduction of spectral information
G. Traven, G. Matijevi\v{c}, T. Zwitter, M. \v{Z}erjal, J. Kos, M., Asplund, J. Bland-Hawthorn, A. R. Casey, G. De Silva, K. Freeman, J. Lin, S., L. Martell, K. J. Schlesinger, S. Sharma, J. D. Simpson, D. B. Zucker, B., Anguiano, G. Da Costa, L. Duong, J. Horner, E. A. Hyde

TL;DR
This paper introduces a semi-automated classification method using t-SNE to identify and categorize peculiar stellar spectra in the Galah survey, enhancing data quality and discovery potential.
Contribution
It applies t-SNE dimensionality reduction to spectral data for effective classification of peculiar spectra in a large stellar survey.
Findings
Majority of spectra are normal stars
Identified 10 categories of peculiar spectra
Catalogue of classified spectra for follow-up studies
Abstract
Galah is an ongoing high-resolution spectroscopic survey with the goal of disentangling the formation history of the Milky Way, using the fossil remnants of disrupted star formation sites which are now dispersed around the Galaxy. It is targeting a randomly selected, magnitude limited () sample of stars, with the goal of observing one million objects. To date, 300,000 spectra have been obtained. Not all of them are correctly processed by parameter estimation pipelines and we need to know about them. We present a semi-automated classification scheme which identifies different types of peculiar spectral morphologies, in an effort to discover and flag potentially problematic spectra and thus help to preserve the integrity of the survey's results. To this end we employ a recently developed dimensionality reduction technique t-SNE (t-distributed Stochastic Neighbour Embedding),…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
