X-ray Reverberation Mapping of Ark 564 using Gaussian Process Regression

TL;DR

This paper uses Gaussian process regression and relativistic reverberation modeling on X-ray data from Ark 564 to measure black hole mass and explore AGN timing, demonstrating novel application of machine learning in this context.

Contribution

It introduces the first use of multi-task learning with Gaussian processes for Fourier-resolved timing analysis in AGN reverberation studies.

Findings

Constrained the black hole mass to approximately 2.3 million solar masses.

Detected a strong, relativistically broadened iron line.

Highlighted the need for additional components to model the soft excess.

Abstract

Ark 564 is an extreme high-Eddington Narrow-line Seyfert 1 galaxy, known for being one of the brightest, most rapidly variable soft X-ray AGN, and for having one of the lowest temperature coronae. Here we present a 410-ks NuSTAR observation and two 115-ks XMM-Newton observations of this unique source, which reveal a very strong, relativistically broadened iron line. We compute the Fourier-resolved time lags by first using Gaussian processes to interpolate the NuSTAR gaps, implementing the first employment of multi-task learning for application in AGN timing. By fitting simultaneously the time lags and the flux spectra with the relativistic reverberation model RELTRANS, we constrain the mass at $2.3^{+2.6}_{-1.3} \times 10^6M_\odot$, although additional components are required to describe the prominent soft excess in this source. These results motivate future combinations of machine learning, Fourier-resolved timing, and the development of reverberation models.