Surveying the reach and maturity of machine learning and artificial intelligence in astronomy

Abstract

Machine learning (automated processes that learn by example in order to classify, predict, discover, or generate new data) and artificial intelligence (methods by which a computer makes decisions or discoveries that would usually require human intelligence) are now firmly established in astronomy. Every week, new applications of machine learning and artificial intelligence are added to a growing corpus of work. Random forests, support vector machines, and neural networks are now having a genuine impact for applications as diverse as discovering extrasolar planets, transient objects, quasars, and gravitationally lensed systems, forecasting solar activity, and distinguishing between signals and instrumental effects in gravitational wave astronomy. This review surveys contemporary, published literature on machine learning and artificial intelligence in astronomy and astrophysics. Applications span seven main categories of activity: classification, regression, clustering, forecasting, generation, discovery, and the development of new scientific insights. These categories form the basis of a hierarchy of maturity, as the use of machine learning and artificial intelligence emerges, progresses, or becomes established. This article is categorized under: Application Areas > Science and Technology Fundamental Concepts of Data and Knowledge > Motivation and Emergence of Data Mining Technologies > Machine Learning

Machine learning technique for morphological classification of galaxies from the SDSS

“…An excellent introduction to the classification algorithms for astronomical tasks, including the morphological galaxy classification, is given in various studies (Ball & Brunner (2010); Way et al (2012); VanderPlas et al (2012); Ivezic & Babu (2014); Al-Jarrah et al (2015); Fluke & Jacobs (2020); El Bouchefry & de Souza (2020), and Vavilova et al (2020). We also refer to the classical work by Buta (2011), and to a good pedagogical review by Conselice et al (2014) with a discussion of principal methods in which galaxies are studied morphologically and structurally.…”

Photometric Redshifts With Machine Learning, Lights and Shadows on a Complex Data Science Use Case

“…In general, the ML-based techniques are able to produce a high-quality photo-z estimation within the photometric ranges imposed by the spectroscopic training set but are less capable of reaching the same photo-z estimation quality outside those ranges. Nevertheless, the positive contribution of data-driven methodologies to the estimation of distances for galaxies and peculiar objects, such as quasars (Baron, 2019;Fluke and Jacobs, 2020), is well-known. Without claiming to be exhaustive, we can cite the following methods proposed in the literature, which testify to their diversity of approach:…”

Visualization in Astrophysics: Developing New Methods, Discovering Our Universe, and Educating the Earth

“…During our recent astrophysics workshops, many astronomers voiced their desires as well as concerns about using machine learning techniques (ML), and in particular, deep learning in the astrophysics discovery processes, mostly surrounding the interpretability of “black box” ML models. Active discussions have concerned the maturity of ML in astronomy, and a number of surveys have been created to assess this maturity [BB10, FJ20, NAB∗19].…”