Light curve fingerprints: an automated approach to the extraction of X-ray variability patterns with feature aggregation -- an example application to GRS 1915+105

Abstract

Time series data mining is an important field of research in the era of "Big Data". Next generation astronomical surveys will generate data at unprecedented rates, creating the need for automated methods of data analysis. We propose a method of light curve characterisation that employs a pipeline consisting of a neural network with a Long-Short Term Memory Variational Autoencoder architecture and a Gaussian mixture model. The pipeline performs extraction and aggregation of features from light curve segments into feature vectors of fixed length which we refer to as light curve "fingerprints". This representation can be readily used as input of down-stream machine learning algorithms. We demonstrate the proposed method on a data set of Rossi X-ray Timing Explorer observations of the galactic black hole X-ray binary GRS 1915+105, which was chosen because of its observed complex X-ray variability. We find that the proposed method can generate a representation that characterises the observations and reflects the presence of distinct classes of GRS 1915+105 X-ray flux variability. We find that this representation can be used to perform efficient classification of light curves. We also present how the representation can be used to quantify the similarity of different light curves, highlighting the problem of the popular classification system of GRS 1915+105 observations, which does not account for intermediate class behaviour.

References in corpus (41)

• SciPy 1.0--Fundamental Algorithms for Scientific Computing in Python
• LSST: from Science Drivers to Reference Design and Anticipated Data Products
• Deep learning for time series classification: a review
• The Swift X-ray Telescope
• cuDNN: Efficient Primitives for Deep Learning
• The MAXI Mission on the ISS: Science and Instruments for Monitoring All Sky X-Ray Images
• GRS 1915+105 and the Disc-Jet Coupling in Accreting Black Hole Systems
• Identifying Exoplanets with Deep Learning: A Five Planet Resonant Chain around Kepler-80 and an Eighth Planet around Kepler-90
• A First Catalog of Variable Stars Measured by the Asteroid Terrestrial-impact Last Alert System (ATLAS)
• ULTRACAM: an ultra-fast, triple-beam CCD camera for high-speed astrophysics
• On Machine-Learned Classification of Variable Stars with Sparse and Noisy Time-Series Data
• An unstable central disk in the superluminal black-hole X-ray binary GRS 1915+105
• A unified model for the spectral variability in GRS 1915+105
• Hard X-ray states and radio emission in GRS 1915+105
• A recurrent neural network for classification of unevenly sampled variable stars
• Quasi periodic oscillations in black hole binaries
• Time-Dependent Disk Models for the Microquasar GRS1915+105
• K2 Variable Catalogue II: Machine Learning Classification of Variable Stars and Eclipsing Binaries in K2 Fields 0-4
• Deep Recurrent Neural Networks for Supernovae Classification
• Characterizing a new class of variability in GRS 1915+105 with simultaneous INTEGRAL/RXTE observations
• Inclination-dependent spectral and timing properties in transient black hole X-ray binaries
• Deep-Learnt Classification of Light Curves
• Scalable End-to-end Recurrent Neural Network for Variable star classification
• High-Frequency Quasi-Periodic Oscillations from GRS 1915+105
• The Host Galaxies of Rapidly Evolving Transients in the Dark Energy Survey
• Clustering Based Feature Learning on Variable Stars
• Unsupervised Classification of Variable Stars
• An Atlas of Exotic Variability in IGR J17091-3624: A Comparison with GRS 1915+105
• Exploring the Long-Term Evolution of GRS 1915+105
• Spectral states of the X-ray binary IGR J17091--3624 observed by INTEGRAL and RXTE
• First INTEGRAL observations of GRS 1915+105
• X-ray properties of the microquasar GRS 1915+105 during a variability class transition
• Discovery of GRS 1915+105 variability patterns in the Rapid Burster
• Nonlinear time series anaysis of the light curves from the black hole system GRS1915+105
• Classification of Periodic Variable Stars with Novel Cyclic-Permutation Invariant Neural Networks
• HiPERCAM: A high-speed, quintuple-beam CCD camera for the study of rapid variability in the Universe
• A non-linear mathematical model for the X-ray variability classes of the microquasar GRS 1915+105 -- II: transition and swaying classes
• A Machine Learning Approach For Classifying Low-mass X-ray Binaries Based On Their Compact Object Nature
• OPTICAM: A triple-camera optical system designed to explore the fastest timescales in Astronomy
• The Mystery of Photometric Twins DES17X1boj and DES16E2bjy
• An Algorithm for the Visualization of Relevant Patterns in Astronomical Light Curves

Cited by in corpus (1)

• Mapping the X-ray variability of GRS1915+105 with machine learning