AGN X-ray Spectroscopy with Neural Networks

Abstract

We explore the possibility of using machine learning to estimate physical parameters directly from AGN X-ray spectra without needing computationally expensive spectral fitting. Specifically, we consider survey quality data, rather than long pointed observations, to ensure that this approach works in the regime where it is most likely to be applied. We simulate Athena WFI spectra of AGN with warm absorbers, and train simple neural networks to estimate the ionisation and column density of the absorbers. We find that this approach can give comparable accuracy to spectral fitting, without the risk of outliers caused by the fit sticking in a false minimum, and with an improvement of around three orders of magnitude in speed. We also demonstrate that using principal component analysis to reduce the dimensionality of the data prior to inputting it into the neural net can significantly increase the accuracy of the parameter estimation for negligible computational cost, while also allowing a simpler network architecture to be used.

arXiv comment8 pages, 4 figures, accepted for publication in MNRAS

References in corpus (21)

• On the Absorption of X-rays in the Interstellar Medium
• Rotation-invariant convolutional neural networks for galaxy morphology prediction
• An X-ray spectral study of 24 type-1 AGN
• A rapidly spinning supermassive black hole at the centre of NGC 1365
• Black hole feedback in the luminous quasar PDS 456
• A fast and long-lived outflow from the supermassive black hole in NGC 5548
• Star-galaxy Classification Using Deep Convolutional Neural Networks
• Estimating photometric redshifts with artificial neural networks
• Warm absorbers in X-rays (WAX), a comprehensive high resolution grating spectral study of a sample of Seyfert galaxies: I. A global view and frequency of occurrence of warm absorbers
• Classifying Radio Galaxies with Convolutional Neural Network
• Photometric redshifts with the Multilayer Perceptron Neural Network: application to the HDF-S and SDSS
• Selection of radio pulsar candidates using artificial neural networks
• XMM-NEWTON High Resolution Spectroscopy of NGC 5548
• An Artificial Neural Network Approach to Classification of Galaxy Spectra
• SPECULATOR: Emulating stellar population synthesis for fast and accurate galaxy spectra and photometry
• Revealing the X-ray Variability of AGN with Principal Component Analysis
• Rapid Classification of TESS Planet Candidates with Convolutional Neural Networks
• The nuclear environment of the NLS1 Mrk 335: obscuration of the X-ray line emission by a variable outflow
• Principal Component Analysis of MCG--6-30-15 with XMM-Newton
• The X-ray Disk/Wind Degeneracy in AGN
• Neural network-based preprocessing to estimate the parameters of the X-ray emission of a single-temperature thermal plasma

Cited by in corpus (6)

• Physics Beyond the Standard Model with Future X-ray Observatories: Projected Constraints on Very-Light Axion-Like Particles with $Athena$ and $AXIS$
• Relativistic Reflection Modeling in AGN and Related Variability from PCA: A Brief Review
• Comparing Reflection and Absorption Models for the Soft X-ray Variability in the NLS1 AGN UGC 11763
• Rapid Spectral Parameter Prediction for Black Hole X-Ray Binaries using Physicalised Autoencoders
• Quantum Machine Learning for Identifying Transient Events in X-ray Light Curves
• X-ray spectral fitting with Monte Carlo Dropout Neural Networks