Everything we’d like to do with LSST data, but we don’t know (yet) how

During the last decade, considerable effort has been made to perform automatic classification of variable stars using machine learning techniques. Traditionally, light curves are represented as a vector of descriptors or features used as input for many algorithms. Some features are computationally expensive, cannot be updated quickly and hence for large datasets such as the LSST cannot be applied. Previous work has been done to develop alternative unsupervised feature extraction algorithms for light curves, but the cost of doing so still remains high. In this work, we propose an end-to-end algorithm that automatically learns the representation of light curves that allows an accurate automatic classification. We study a series of deep learning architectures based on Recurrent Neural Networks and test them in automated classification scenarios. Our method uses minimal data preprocessing, can be updated with a low computational cost for new observations and light curves, and can scale up to massive datasets. We transform each light curve into an input matrix representation whose elements are the differences in time and magnitude, and the outputs are classification probabilities. We test our method in three surveys: OGLE-III, Gaia and WISE. We obtain accuracies of about 95% in the main classes and 75% in the majority of subclasses. We compare our results with the Random Forest classifier and obtain competitive accuracies while being faster and scalable. The analysis shows that the computational complexity of our approach grows up linearly with the light curve size, while the traditional approach cost grows as N log (N).

show abstract

“…The technological challenges imposed by the LSST are plenty (Ivezić et al 2016). One of those challenges is the alerts and broker systems.…”

Section: Introductionmentioning

confidence: 99%

Scalable end-to-end recurrent neural network for variable star classification

Becker

Pichara

Catelan

et al. 2020

Monthly Notices of the Royal Astronomical Society

View full text Add to dashboard Cite

show abstract

“…Timescales relevant to astronomical phenomena range from milliseconds to centuries, depending on the object type, and this hints at major challenges in both data acquisition and data management. The LSST project, one of many facing these challenges, is leveraging the experience of large particle-physics experiments to develop the necessary sophisticated computational architecture [26].…”

Section: Data Processing and Measurementsmentioning

confidence: 99%

“…The Megacam instrument on the Canada-France-Hawaii Telescope, commissioned in 2003, has 340 megapixels and produces a typical raw data volume of ∼ 100 GB per night [24]. The camera on the future Large Synoptic Survey Telescope (LSST) will have over 3 gigapixels and produce tens of terabytes of raw data per night [26].…”

Section: Capturing the Light: Instruments And Detectorsmentioning

confidence: 99%

Astronomical observations: a guide for allied researchers

Barmby¹

2019

TheOJA

View full text Add to dashboard Cite

Observational astrophysics uses sophisticated technology to collect and measure electromagnetic and other radiation from beyond the Earth. Modern observatories produce large, complex datasets and extracting the maximum possible information from them requires the expertise of specialists in many fields beyond physics and astronomy, from civil engineers to statisticians and software engineers. This article introduces the essentials of professional astronomical observations to colleagues in allied fields, to provide context and relevant background for both facility construction and data analysis. It covers the path of electromagnetic radiation through telescopes, optics, detectors, and instruments, its transformation through processing into measurements and information, and the use of that information to improve our understanding of the physics of the cosmos and its history.

show abstract

“…The next generation of imaging surveys, e.g., Large Synoptic Survey Telescope (LSST Science Collaboration et al 2009) and Chinese Space Station Telescope (Zhan 2021), will yield large amounts of high-quality data. Maximizing the extraction of useful and reliable information from the huge amount of data is one of the goals pursued by the next generation of sky surveys (Ivezić et al 2016). Machine learning offers a promising approach for extracting important features and maximizing the use of astronomical data.…”

Section: Introductionmentioning

confidence: 99%

L-dwarf Detection from SDSS Images using Improved Faster R-CNN

Cao

Pan

et al. 2023

View full text Add to dashboard Cite

We present a data-driven approach to automatically detect L dwarfs from Sloan Digital Sky Survey (SDSS) images using an improved Faster R-CNN framework based on deep learning. The established L-dwarf automatic detection (LDAD) model distinguishes L dwarfs from other celestial objects and backgrounds in SDSS field images by learning the features of 387 SDSS images containing L dwarfs. Applying the LDAD model to the SDSS images containing 93 labeled L dwarfs in the test set, we successfully detected 83 known L dwarfs with a recall rate of 89.25% for known L dwarfs. Several techniques are implemented in the LDAD model to improve its detection performance for L dwarfs, including the deep residual network and the feature pyramid network. As a result, the LDAD model outperforms the model of the original Faster R-CNN, whose recall rate of known L dwarfs is 80.65% for the same test set. The LDAD model was applied to detect L dwarfs from a larger validation set including 843 labeled L dwarfs, resulting in a recall rate of 94.42% for known L dwarfs. The newly identified candidates include L dwarfs, late M and T dwarfs, which were estimated from color (i − z) and spectral type relation. The contamination rates for the test candidates and validation candidates are 8.60% and 9.27%, respectively. The detection results indicate that our model is effective to search for L dwarfs from astronomical images.

show abstract

Everything we’d like to do with LSST data, but we don’t know (yet) how

Cited by 11 publications

References 16 publications

Scalable end-to-end recurrent neural network for variable star classification

Scalable end-to-end recurrent neural network for variable star classification

Astronomical observations: a guide for allied researchers

L-dwarf Detection from SDSS Images using Improved Faster R-CNN

Contact Info

Product

Resources

About