Rapid Spectral Parameter Prediction for Black Hole X-Ray Binaries using Physicalised Autoencoders

Abstract

Black hole X-ray binaries (BHBs) offer insights into extreme gravitational environments and the testing of general relativity. The X-ray spectrum collected by NICER offers valuable information on the properties and behaviour of BHBs through spectral fitting. However, traditional spectral fitting methods are slow and scale poorly with model complexity. This paper presents a new semi-supervised autoencoder neural network for parameter prediction and spectral reconstruction of BHBs, showing an improvement of up to a factor of 2,700 in speed while maintaining comparable accuracy. The approach maps the spectral features from the numerous outbursts catalogued by NICER and generalises them to new systems for efficient and accurate spectral fitting. The effectiveness of this approach is demonstrated in the spectral fitting of BHBs and holds promise for use in other areas of astronomy and physics for categorising large datasets.

arXiv comment13 pages, 13 figures

References in corpus (24)

• On the Absorption of X-rays in the Interstellar Medium
• WaveNet: A Generative Model for Raw Audio
• X-ray Properties of Black-Hole Binaries
• The James Webb Space Telescope
• Advection-Dominated Accretion and the Spectral States of Black Hole X-Ray Binaries: Application to Nova Muscae 1991
• Modelling the behaviour of accretion flows in X-ray binaries
• On the Properties of Neural Machine Translation: Encoder-Decoder Approaches
• A review of quasi-periodic oscillations from black hole X-ray binaries: observation and theory
• A Global Spectral Study of Black Hole X-ray Binaries
• A Simple Comptonization Model
• Mass Measurements of Stellar and Intermediate Mass Black-Holes
• An Empirical Background Model for the NICER X-ray Timing Instrument
• Testing Accretion Disk Theory in Black Hole X-ray Binaries
• Evidence for Frame-Dragging Around Spinning Black Holes in X-Ray Binaries
• Testing the space-time geometry around black hole candidates with the analysis of the broad K$\alpha$ iron line
• Multi-Temperature Blackbody Spectra of Thin Accretion Disks With and Without a Zero-Torque Inner Boundary Condition
• On the determination of the spin and disc truncation of accreting black holes using X-ray reflection
• Self-Consistent Black Hole Accretion Spectral Models and the Forgotten Role of Coronal Comptonization of Reflection Emission
• Re-estimating the Spin Parameter of the Black Hole in Cygnus X-1
• Schwarzschild and Kerr Solutions of Einstein's Field Equation -- an introduction
• Line Emission Mapper (LEM): Probing the physics of cosmic ecosystems
• The Importance of Being Interpretable: Toward An Understandable Machine Learning Encoder for Galaxy Cluster Cosmology
• AGN X-ray Spectroscopy with Neural Networks
• The Wide Field Imager Instrument for Athena

Cited by in corpus (1)

• Extracting latent representations from X-ray spectra. Classification, regression, and accretion signatures of Chandra sources