A review of unsupervised learning in astronomy

Abstract

This review summarizes popular unsupervised learning methods, and gives an overview of their past, current, and future uses in astronomy. Unsupervised learning aims to organise the information content of a dataset, in such a way that knowledge can be extracted. Traditionally this has been achieved through dimensionality reduction techniques that aid the ranking of a dataset, for example through principal component analysis or by using auto-encoders, or simpler visualisation of a high dimensional space, for example through the use of a self organising map. Other desirable properties of unsupervised learning include the identification of clusters, i.e. groups of similar objects, which has traditionally been achieved by the k-means algorithm and more recently through density-based clustering such as HDBSCAN. More recently, complex frameworks have emerged, that chain together dimensionality reduction and clustering methods. However, no dataset is fully unknown. Thus, nowadays a lot of research has been directed towards self-supervised and semi-supervised methods that stand to gain from both supervised and unsupervised learning.

30 pages, 6 figures. Invited contribution to special issue in Astronomy & Computing

References in corpus (60)

• The Cosmic Evolution Survey (COSMOS) -- Overview
• K-corrections and filter transformations in the ultraviolet, optical, and near infrared
• TEMPO2, a new pulsar timing package. I: Overview
• Machine Learning in Astronomy: a practical overview
• Hunting for open clusters in \textit{Gaia} DR2: the Galactic anticentre
• The weirdest SDSS galaxies: results from an outlier detection algorithm
• Estimating stellar atmospheric parameters, absolute magnitudes and elemental abundances from the LAMOST spectra with Kernel-based Principal Component Analysis
• An automatic taxonomy of galaxy morphology using unsupervised machine learning
• Galaxy morphological classification in deep-wide surveys via unsupervised machine learning
• Astronomia ex machina: a history, primer, and outlook on neural networks in astronomy
• Modelling type 1 quasar colours in the era of Rubin and Euclid
• Data Mining for Gravitationally Lensed Quasars
• Unsupervised star, galaxy, qso classification: Application of HDBSCAN
• Cataloging the radio-sky with unsupervised machine learning: a new approach for the SKA era
• The GOGREEN survey: Post-infall environmental quenching fails to predict the observed age difference between quiescent field and cluster galaxies at z>1
• Systematic Serendipity: A Test of Unsupervised Machine Learning as a Method for Anomaly Detection
• Uncloaking hidden repeating fast radio bursts with unsupervised machine learning
• The narrow-line region of narrow-line and broad-line type 1 AGN I. A zone of avoidance in density
• Unveiling the rarest morphologies of the LOFAR Two-metre Sky Survey radio source population with self-organised maps
• Denoising Gravitational Waves with Enhanced Deep Recurrent Denoising Auto-Encoders
• Unveiling hidden stellar aggregates in the Milky Way: 1656 new star clusters found in Gaia EDR3
• Unsupervised machine learning for transient discovery in Deeper, Wider, Faster light curves
• Machine Learning of Interstellar Chemical Inventories
• Deep Neural Network Classifier for Variable Stars with Novelty Detection Capability
• Deep Learning Assisted Data Inspection for Radio Astronomy
• Hunting for C-rich long-period variable stars in the Milky Way's bar-bulge using unsupervised classification of Gaia BP/RP spectra
• Unsupervised Galaxy Morphological Visual Representation with Deep Contrastive Learning
• Discovery of Peculiar Radio Morphologies with ASKAP using Unsupervised Machine Learning
• Unsupervised feature-learning for galaxy SEDs with denoising autoencoders
• Autoencoding Galaxy Spectra II: Redshift Invariance and Outlier Detection
• Automatic Survey-Invariant Variable Star Classification
• The membership of stars, density profile and mass segregation in open clusters using a new machine learning-based method
• SKA Science Data Challenge 2: analysis and results
• A Robust Study of High-Redshift Galaxies: Unsupervised Machine Learning for Characterising morphology with JWST up to z ~ 8
• Mapping the Diversity of Galaxy Spectra with Deep Unsupervised Machine Learning
• Classification of OGLE eclipsing binary stars based on their morphology type with Locally Linear Embedding
• ASTErIsM - Application of topometric clustering algorithms in automatic galaxy detection and classification
• Massive young stellar objects in the Local Group irregular galaxy NGC6822 identified using machine learning
• A Gaia astrometric view of the open clusters Pleiades, Praesepe and Blanco 1
• Cleaning foregrounds from single-dish 21cm intensity maps with Kernel Principal Component Analysis
• The Deeper, Wider, Faster Program: Exploring stellar flare activity with deep, fast cadenced DECam imaging via machine learning
• Physics-aware Machine Learning Revolutionizes Scientific Paradigm for Machine Learning and Process-based Hydrology
• Possible Evidence of a Universal Radio/X-ray Correlation in a near-Complete Sample of Hard X-ray Selected Seyfert Galaxies
• Galaxy And Mass Assembly (GAMA): Self-Organizing Map Application on Nearby Galaxies
• X-ray Study of Spatial Structures in Tycho's Supernova Remnant Using Unsupervised Deep Learning
• Inner and outer rings are not strongly coupled with stellar bars
• Mapping the X-ray variability of GRS1915+105 with machine learning
• Tidal debris from Omega Centauri discovered with unsupervised machine learning
• Milky Way-like galaxies: stellar population properties of dynamically defined disks, bulges and stellar halos
• Classifying a frequently repeating fast radio burst, FRB 20201124A, with unsupervised machine learning
• Unsupervised spectral decomposition of X-ray binaries with application to GX 339-4
• Mapping the Similarities of Spectra: Global and Locally-biased Approaches to SDSS Galaxy Data
• Classification of local ultraluminous infrared galaxies and quasars with kernel principal component analysis
• Self-Supervised Clustering on Image-Subtracted Data with Deep-Embedded Self-Organizing Map
• Learning Reionization History from Quasars with Simulation-Based Inference
• AstronomicAL: An interactive dashboard for visualisation, integration and classification of data using Active Learning
• DeepAstroUDA: Semi-Supervised Universal Domain Adaptation for Cross-Survey Galaxy Morphology Classification and Anomaly Detection
• Unsupervised Method for Correlated Noise Removal for Multi-wavelength Exoplanet Transit Observations
• Interpreting Stellar Spectra with Unsupervised Domain Adaptation
• Gravitational Dimensionality Reduction Using Newtonian Gravity and Einstein's General Relativity

Cited by in corpus (12)

• Wide Area VISTA Extra-galactic Survey (WAVES): Unsupervised star-galaxy separation on the WAVES-Wide photometric input catalogue using UMAP and $\mathrm{HDBSCAN}$
• Unsupervised Machine Learning for Classifying CHIME Fast Radio Bursts and Investigating Empirical Relations
• Explainable autoencoder for neutron star dense matter parameter estimation
• Which is which? Identification of the two compact objects in gravitational-wave binaries
• Semi-supervised classification of stars, galaxies and quasars using K-means and random-forest approaches
• How to set up your first machine learning project in astronomy
• Navigating AGN variability with self-organizing maps
• Identifying Radio Active Galactic Nuclei with Machine Learning and Large-Area Surveys
• Deblending the MIGHTEE-COSMOS survey with XID+: The resolved radio source counts to $S_{1.4}\approx 5\mu$Jy
• Deciphering galaxy images using machine vision -- Combining variational autoencoder and principal component analysis for feature extraction
• Extracting latent representations from X-ray spectra. Classification, regression, and accretion signatures of Chandra sources
• Estimating the peak energy of Swift gamma-ray bursts using supervised machine learning