arXiv subjects

James Hung-Hsu Chan

Publications and source records attributed to James Hung-Hsu Chan.

8 recordsLinked to original sources

arXiv2026

Microcaustics: Fast GPU-Accelerated Dynamical Microlensing of Time-Dependent Extended Sources

Gravitational microlensing probes unresolved quasar accretion disc structure and compact objects in lens galaxies and is an important systematic in strong-lens time-delay cosmography. Most analyses use static microlensing map simulations because evolving individual microlenses and caustics is computationally prohibitive. We present Microcaustics, a GPU-accelerated Python package built on PyTorch for dynamical microlensing of time-dependent extended sources in strong gravitational lenses. It combines a local-exact/far-field deflection solver, source-directed lens-plane scouting with temporal reuse, and biquadratic inverse polygon mapping (IPM). GPU batching and fused Triton rasterization accelerate the calculation, while binary caustic labels track source-centred crossings. Across three simulated quadruply lensed quasar systems, our package generates a 10-year, 147-frame multiband light curve in an amortized median of 0.68 s (0.88 s with labels) on a consumer GPU with a median light-curve RMSE of 0.35 mmag, while recovering every reference source-centred caustic crossing. At comparable map accuracy, the batched throughput is 780 times higher than that of a recent CUDA IPM implementation and extrapolates to a speedup of millions relative to classic GPU-D. We demonstrate that our IPM converges as $N_{\rm ray}^{-0.84}$, approaching the idealized $N_{\rm ray}^{-1}$ convergence rate of piecewise-affine IPM and improving upon the $N_{\rm ray}^{-1/2}$ inverse ray-shooting rate. We also introduce a GPU-resident general relativistic and reverberating quasar source model and demonstrate our method with an evolving supernova source. These advances make fully dynamical stellar fields practical for Bayesian and simulation-based inference.

astro-ph.GA↗

arXiv2025

Joint Modeling of Quasar Variability and Accretion Disk Reprocessing using Latent Stochastic Differential Equations

Quasars are bright active galactic nuclei powered by the accretion of matter around supermassive black holes at the center of galaxies. Their stochastic brightness variability depends on the physical properties of the accretion disk and black hole. The upcoming Rubin Observatory Legacy Survey of Space and Time (LSST) is expected to observe tens of millions of quasars, so there is a need for efficient techniques like machine learning that can handle the large volume of data. Quasar variability is believed to be driven by an X-ray corona, which is reprocessed by the accretion disk and emitted as UV/optical variability. We are the first to introduce an auto-differentiable simulation of the accretion disk and reprocessing. We use the simulation as a direct component of our neural network to jointly model the driving variability and reprocessing, trained with supervised learning on simulated LSST-like 10 yr quasar light curves. We encode the light curves using a transformer encoder, and the driving variability is reconstructed using latent stochastic differential equations, a physically motivated generative deep learning method that can model continuous-time stochastic dynamics. By embedding the physical processes of the driving signal and reprocessing into our network, we achieve a model that is more robust and interpretable. We demonstrate that our model outperforms a Gaussian process regression baseline and can infer accretion disk parameters and time delays between wave bands, even for out-of-distribution driving signals. Our approach provides a powerful framework that can be adapted to solve other inverse problems in multivariate time series.

astro-ph.GA↗

arXiv2024

Reverberation Mapping of Lamp-post and Wind Structures in Accretion Thin Disks

To address the discrepancy where disk sizes exceed those predicted by standard models, we explore two extensions to disk size estimates within the UV/optical wavelength range: disk winds and color correction. We provide detailed, self-consistent derivations and analytical formulas, including those based on a power-law temperature approximation, offering efficient tools for analyzing observational data. Applying our model to four type I AGNs with intensive reverberation mapping observations, we find a shallower temperature slope ($T\propto R^{-0.66}$, compared to $R^{-3/4}$ traditionally) and a color correction factor ($f_{\rm col} \approx 1.6$), consistent with previous studies. We observe a positive correlation between accretion rate and color correction with black hole mass. However, the small sample size limits our conclusions. The strong degeneracy between the temperature slope and accretion rate suggests that incorporating flux spectra or spectral energy distributions could improve fitting accuracy. Our simulation approach rapidly generates quasar light curves while accommodating various observational scenarios for reverberation mapping, making it well-suited for training machine learning algorithms.

astro-ph.GA↗

arXiv2024

Latent Stochastic Differential Equations for Modeling Quasar Variability and Inferring Black Hole Properties

Quasars are bright and unobscured active galactic nuclei (AGN) thought to be powered by the accretion of matter around supermassive black holes at the centers of galaxies. The temporal variability of a quasar's brightness contains valuable information about its physical properties. The UV/optical variability is thought to be a stochastic process, often represented as a damped random walk described by a stochastic differential equation (SDE). Upcoming wide-field telescopes such as the Rubin Observatory Legacy Survey of Space and Time (LSST) are expected to observe tens of millions of AGN in multiple filters over a ten year period, so there is a need for efficient and automated modeling techniques that can handle the large volume of data. Latent SDEs are machine learning models well suited for modeling quasar variability, as they can explicitly capture the underlying stochastic dynamics. In this work, we adapt latent SDEs to jointly reconstruct multivariate quasar light curves and infer their physical properties such as the black hole mass, inclination angle, and temperature slope. Our model is trained on realistic simulations of LSST ten year quasar light curves, and we demonstrate its ability to reconstruct quasar light curves even in the presence of long seasonal gaps and irregular sampling across different bands, outperforming a multioutput Gaussian process regression baseline. Our method has the potential to provide a deeper understanding of the physical properties of quasars and is applicable to a wide range of other multivariate time series with missing data and irregular sampling.

astro-ph.GA↗

arXiv2023

LenSiam: Self-Supervised Learning on Strong Gravitational Lens Images

Self-supervised learning has been known for learning good representations from data without the need for annotated labels. We explore the simple siamese (SimSiam) architecture for representation learning on strong gravitational lens images. Commonly used image augmentations tend to change lens properties; for example, zoom-in would affect the Einstein radius. To create image pairs representing the same underlying lens model, we introduce a lens augmentation method to preserve lens properties by fixing the lens model while varying the source galaxies. Our research demonstrates this lens augmentation works well with SimSiam for learning the lens image representation without labels, so we name it LenSiam. We also show that a pre-trained LenSiam model can benefit downstream tasks. We open-source our code and datasets at https://github.com/kuanweih/LenSiam .

astro-ph.IM↗

arXiv2023

Condensation and Evaporation of Boson Stars

Axion-like particles, including the QCD axion, are well-motivated dark matter candidates. Numerical simulations have revealed coherent soliton configurations, also known as boson stars, in the centers of axion halos. We study evolution of axion solitons immersed into a gas of axion waves with Maxwellian velocity distribution. Combining analytical approach with controlled numerical simulations we find that heavy solitons grow by condensation of axions from the gas, while light solitons evaporate. We deduce the parametric dependence of the soliton growth/evaporation rate and show that it is proportional to the rate of the kinetic relaxation in the gas. The proportionality coefficient is controlled by the product of the soliton radius and the typical gas momentum or, equivalently, the ratio of the gas and soliton virial temperatures. We discuss the asymptotics of the rate when this parameter is large or small.

astro-ph.CO↗

arXiv2023

Boson Star Normal Modes

Boson stars are gravitationally bound objects that arise in ultralight dark matter models and form in the centers of galactic halos or axion miniclusters. We systematically study the excitations of a boson star, taking into account the mixing between positive and negative frequencies introduced by gravity. We show that the spectrum contains zero-energy modes in the monopole and dipole sectors resulting from spontaneous symmetry breaking by the boson star background. We analyze the general properties of the eigenmodes and derive their orthogonality and completeness conditions which have non-standard form due to the positive-negative frequency mixing. The eigenvalue problem is solved numerically for the first few energy levels in different multipole sectors and the results are compared to the solutions of the Schrödinger equation in fixed boson star gravitational potential. The two solutions differ significantly for the lowest modes, but get close for higher levels. We further confirm the normal mode spectrum in 3D wave simulations where we inject perturbations with different multipoles. As an application of the normal mode solutions, we compute the matrix element entering the evaporation rate of a boson star immersed in a hot axion gas. The computation combines the use of exact wavefunctions for the low-lying bound states and of the Schrödinger approximation for the high-energy excitations.

astro-ph.CO↗

arXiv2022

A Highly Magnified Gravitationally Lensed Red QSO at z = 2.5 with a Significant Flux Ratio Anomaly

We present the discovery of a gravitationally lensed dust-reddened QSO at z = 2.517, identified in a survey for QSOs by infrared selection. Hubble Space Telescope imaging reveals a quadruply lensed system in a cusp configuration, with a maximum image separation of ~1.8\arcsec. We find that compared to the central image of the cusp, the neighboring brightest image is anomalous by a factor of ~ 7 - 10, which is the largest flux anomaly measured to date in a lensed QSO. Incorporating high-resolution Jansky Very Large Array radio imaging and sub-mm imaging with the Atacama Large (sub-)Millimetre Array, we conclude that a low-mass perturber is the most likely explanation for the anomaly. The optical through near-infrared spectrum reveals that the QSO is moderately reddened with E(B - V) = 0.7 - 0.9. We see an upturn in the ultraviolet spectrum due to ~ 1% of the intrinsic emission being leaked back into the line of sight, which suggests that the reddening is intrinsic and not due to the lens. The QSO may have an Eddington ratio as high as L/L_Edd ~ 0.2. Consistent with previous red QSO samples, this source exhibits outflows in its spectrum as well as morphological properties suggestive of it being in a merger-driven transitional phase. We find a host-galaxy stellar mass of log M_*/M_Sun = 11.4, which is higher than the local M_BH vs. M_* relation, but consistent with other high redshift QSOs. When de-magnified, this QSO is at the knee of the luminosity function, allowing for the detailed study of a more typical moderate-luminosity infrared-selected QSO at high redshift.

astro-ph.GA↗