arXiv subjects

Tamás Budavári

Publications and source records attributed to Tamás Budavári.

At least 19 recordsLinked to original sources

arXiv2026

Galactic Archaeology with the Subaru `Ōnohi`ula Prime Focus Spectrograph Strategic Program

The recently commissioned Subaru `Ōnohi`ula Prime Focus Spectrograph (PFS) will obtain spectra from nearly 2,400 fibers that cover 1.24 square degrees. The 360 night Subaru Strategic Program for PFS is dedicating approximately one-third of its allocation (130 nights) to study the structure and evolution of galaxies in the Local Group. This Galactic Archaeological survey has three pillars. (1) We will determine whether the mass density profiles of dwarf galaxies are consistent with cusps, as expected for cold dark matter, or cores, as expected from alternative dark matter theories or baryonic feedback. We will deduce the density profiles as a function of radius from modeling of the full line-of-sight velocity and abundance distributions for six dwarf galaxies. Our total sample will consist of 18,000 member stars to beyond the nominal tidal radius of each system. (2) From measurements of the [alpha/Fe] abundance ratio, we will learn the difference in assembly history of the two most massive galaxies in the Local Group: M31 and the Milky Way. We will observe 30,000 member stars over 45 square degrees of M31's halo and outer disk. (3) We will uncover how the most fragile (outer) part of the Milky Way responded to accretion events both in the distant past (such as Gaia-Sausage Enceladus) and in more recent history (such as the Sagittarius dwarf spheroidal galaxy). To support this study, PFS will provide velocities and metallicities--from which, in combination with photometry, we will deduce ages--for tens of thousands of main-sequence stars out to a Galactocentric distance of ~30 kpc.

astro-ph.GA↗

arXiv2026

Generalization from Low- to Moderate-Resolution Spectra with Neural Networks for Stellar Parameter Estimation: A Case Study with DESI

Cross-survey generalization is a critical challenge in stellar spectral analysis, particularly in cases such as transferring from low- to moderate-resolution surveys. We investigate this problem using pre-trained models, focusing on simple neural networks such as multilayer perceptrons (MLPs), with a case study transferring from LAMOST low-resolution spectra (LRS) to DESI medium-resolution spectra (MRS). Specifically, we pre-train MLPs on either LRS or their embeddings and fine-tune them for application to DESI stellar spectra. We compare MLPs trained directly on spectra with those trained on embeddings derived from transformer-based models (self-supervised foundation models pre-trained for multiple downstream tasks). We also evaluate different fine-tuning strategies, including residual-head fine-tuning, LoRA, and full fine-tuning. We find that MLPs pre-trained on LAMOST LRS achieve strong performance, even without fine-tuning, and that modest fine-tuning with DESI spectra further improves the results. For iron abundance, embeddings from a transformer-based model yield advantages in the metal-rich ([Fe/H] > -1.0) regime, but underperform in the metal-poor regime compared to MLPs trained directly on LRS. We also show that the optimal fine-tuning strategy depends on the specific stellar parameter under consideration. These results highlight that simple pre-trained MLPs can provide competitive cross-survey generalization, while the role of spectral foundation models for cross-survey stellar parameter estimation requires further exploration.

astro-ph.SR↗

arXiv2025

ImageMM: Joint multi-frame image restoration and super-resolution

A key processing step in ground-based astronomy involves combining multiple noisy and blurry exposures to produce an image of the night sky with an improved signal-to-noise ratio. Typically, this is achieved via image coaddition, and can be undertaken such that the resulting night sky image has enhanced spatial resolution. Yet, this task remains a formidable challenge despite decades of advancements. In this paper, we introduce ImageMM: a new framework based on the majorization-minimization (MM) algorithm for joint multi-frame astronomical image restoration and super-resolution. ImageMM uses multiple registered astronomical exposures to produce a nonparametric latent image of the night sky, prior to the atmosphere's impact on the observed exposures. Our framework also features a novel variational approach to compute refined point-spread functions of arbitrary resolution for the restoration and super-resolution procedure. Our algorithms, implemented in TensorFlow, leverage graphics processing unit acceleration to produce latent images in near real time, even when processing high-resolution exposures. We tested ImageMM on Hyper Suprime-Cam (HSC) exposures, which are a precursor of the upcoming imaging data from the Rubin Observatory. The results are encouraging: ImageMM produces sharp latent images, in which spatial features of bright sources are revealed in unprecedented detail (e.g., showing the structure of spiral galaxies), and where faint sources that are usually indistinguishable from the noisy sky background also become discernible, thus pushing the detection limits. Moreover, aperture photometry performed on the HSC pipeline coadd and ImageMM's latent images yielded consistent source detection and flux measurements, thereby demonstrating ImageMM's suitability for cutting-edge photometric studies with state-of-the-art astronomical imaging data.

astro-ph.IM↗

arXiv2025

AstroClearNet: Deep image prior for multi-frame astronomical image restoration

Recovering high-fidelity images of the night sky from blurred observations is a fundamental problem in astronomy, where traditional methods typically fall short. In ground-based astronomy, combining multiple exposures to enhance signal-to-noise ratios is further complicated by variations in the point-spread function caused by atmospheric turbulence. In this work, we present a self-supervised multi-frame method, based on deep image priors, for denoising, deblurring, and coadding ground-based exposures. Central to our approach is a carefully designed convolutional neural network that integrates information across multiple observations and enforces physically motivated constraints. We demonstrate the method's potential by processing Hyper Suprime-Cam exposures, yielding promising preliminary results with sharper restored images.

astro-ph.IM↗

arXiv2024

Uncertainty of line-of-sight velocity measurement of faint stars from low and medium resolution optical spectra

Massively multiplexed spectrographs will soon gather large statistical samples of stellar spectra. The accurate estimation of uncertainties on derived parameters, such as line-of-sight velocity $v_\mathrm{los}$, especially for spectra with low signal-to-noise ratios, is paramount. We generated an ensemble of simulated optical spectra of stars as if they were observed with low- and medium-resolution fiber-fed instruments on an 8-meter class telescope, similar to the Subaru Prime Focus Spectrograph, and determined $v_\mathrm{los}$ by fitting stellar templates to the simulations. We compared the empirical errors of the derived parameters -- calculated from an ensemble of simulations -- to the asymptotic error determined from the Fisher matrix, as well as from Monte Carlo sampling of the posterior probability. We confirm that the uncertainty of $v_\mathrm{los}$ scales with the inverse square root of $S/N$, but also show how this scaling breaks down at low $S/N$ and analyze the error and bias caused by template mismatch. We outline a computationally optimized algorithm to fit multi-exposure data and provide the mathematical model of stellar spectrum fitting that maximizes the so called significance, which allows for calculating the error from the Fisher matrix analytically. We also introduce the effective line count, and provide a scaling relation to estimate the error of $v_\mathrm{los}$ measurement based on the stellar type. Our analysis covers a range of stellar types with parameters that are typical of the Galactic outer disk and halo, together with analogs of stars in M31 and in satellite dwarf spheroidal galaxies around the Milky Way.

astro-ph.IM↗

arXiv2023

Splines 'n Lines: Rest-frame galaxy spectral energy distributions via Bayesian functional data analysis

Survey-based measurements of the spectral energy distributions (SEDs) of galaxies have flux density estimates on badly misaligned grids in rest-frame wavelength. The shift to rest frame wavelength also causes estimated SEDs to have differing support. For many galaxies, there are sizeable wavelength regions with missing data. Finally, dim galaxies dominate typical samples and have noisy SED measurements, many near the limiting signal-to-noise level of the survey. These limitations of SED measurements shifted to the rest frame complicate downstream analysis tasks, particularly tasks requiring computation of functionals (e.g., weighted integrals) of the SEDs, such as synthetic photometry, quantifying SED similarity, and using SED measurements for photometric redshift estimation. We describe a hierarchical Bayesian framework, drawing on tools from functional data analysis, that models SEDs as a random superposition of smooth continuum basis functions (B-splines) and line features, comprising a finite-rank, nonstationary Gaussian process, measured with additive Gaussian noise. We apply this *Splines 'n Lines* (SnL) model to a collection of 678,239 galaxy SED measurements comprising the Main Galaxy Sample from the Sloan Digital Sky Survey, Data Release 17, demonstrating capability to provide continuous estimated SEDs that reliably denoise, interpolate, and extrapolate, with quantified uncertainty, including the ability to predict line features where there is missing data by leveraging correlations between line features and the entire continuum.

astro-ph.IM↗

arXiv2021

GPU-Accelerated Hierarchical Bayesian Inference with Application to Modeling Cosmic Populations: CUDAHM

We describe a computational framework for hierarchical Bayesian inference with simple (typically single-plate) parametric graphical models that uses graphics processing units (GPUs) to accelerate computations, enabling deployment on very large datasets. Its C++ implementation, CUDAHM (CUDA for Hierarchical Models) exploits conditional independence between instances of a plate, facilitating massively parallel exploration of the replication parameter space using the single instruction, multiple data architecture of GPUs. It provides support for constructing Metropolis-within-Gibbs samplers that iterate between GPU-accelerated robust adaptive Metropolis sampling of plate-level parameters conditional on upper-level parameters, and Metropolis-Hastings sampling of upper-level parameters on the host processor conditional on the GPU results. CUDAHM is motivated by demographic problems in astronomy, where density estimation and linear and nonlinear regression problems must be addressed for populations of thousands to millions of objects whose features are measured with possibly complex uncertainties. We describe a thinned latent point process framework for modeling such demographic data. We demonstrate accurate GPU-accelerated parametric conditional density deconvolution for simulated populations of up to 300,000 objects in ~1 hour using a single NVIDIA Tesla K40c GPU. Supplementary material provides details about the CUDAHM API and the demonstration problem.

astro-ph.IM↗

arXiv2021

Probabilistic Association of Transients to their Hosts (PATH)

We introduce a new method to estimate the probability that an extragalactic transient source is associated with a candidate host galaxy. This approach relies solely on simple observables: sky coordinates and their uncertainties, galaxy fluxes and angular sizes. The formalism invokes Bayes' rule to calculate the posterior probability P(O_i|x) from the galaxy prior P(O), observables x, and an assumed model for the true distribution of transients in/around their host galaxies. Using simulated transients placed in the well-studied COSMOS field, we consider several agnostic and physically motivated priors and offset distributions to explore the method sensitivity. We then apply the methodology to the set of 13~fast radio bursts (FRBs) localized with an uncertainty of several arcseconds. Our methodology finds nine of these are securely associated to a single host galaxy, P(O_i|x)>0.95. We examine the observed and intrinsic properties of these secure FRB hosts, recovering similar distributions as previous works. Furthermore, we find a strong correlation between the apparent magnitude of the securely identified host galaxies and the estimated cosmic dispersion measures of the corresponding FRBs, which results from the Macquart relation. Future work with FRBs will leverage this relation and other measures from the secure hosts as priors for future associations. The methodology is generic to transient type, localization error, and image quality. We encourage its application to other transients where host galaxy associations are critical to the science, e.g. gravitational wave events, gamma-ray bursts, and supernovae. We have encoded the technique in Python on GitHub: https://github.com/FRBs/astropath.

astro-ph.HE↗

arXiv2020

Computational Tools for the Spectroscopic Analysis of White Dwarfs

The spectroscopic features of white dwarfs are formed in the thin upper layer of their stellar photosphere. These features carry information about the white dwarf's surface temperature, surface gravity, and chemical composition (hereafter 'labels'). Existing methods to determine these labels rely on complex ab-initio theoretical models which are not always publicly available. Here we present two techniques to determine atmospheric labels from white dwarf spectra: a generative fitting pipeline that interpolates theoretical spectra with artificial neural networks, and a random forest regression model using parameters derived from absorption line features. We test and compare our methods using a large catalog of white dwarfs from the Sloan Digital Sky Survey (SDSS), achieving the same accuracy and negligible bias compared to previous studies. We package our techniques into an open-source Python module 'wdtools' that provides a computationally inexpensive way to determine stellar labels from white dwarf spectra observed from any facility. We will actively develop and update our tool as more theoretical models become publicly available. We discuss applications of our tool in its present form to identify interesting outlier white dwarf systems including those with magnetic fields, helium-rich atmospheres, and double-degenerate binaries.

astro-ph.SR↗

arXiv2020

Optimal Probabilistic Catalogue Matching for Radio Sources

Cross-matching catalogues from radio surveys to catalogues of sources at other wavelengths is extremely hard, because radio sources are often extended, often consist of several spatially separated components, and often no radio component is coincident with the optical/infrared host galaxy. Traditionally, the cross-matching is done by eye, but this does not scale to the millions of radio sources expected from the next generation of radio surveys. We present an innovative automated procedure, using Bayesian hypothesis testing, that models trial radio-source morphologies with putative positions of the host galaxy. This new algorithm differs from an earlier version by allowing more complex radio source morphologies, and performing a simultaneous fit over a large field. We show that this technique performs well in an unsupervised mode.

astro-ph.IM↗

arXiv2019

Robust Registration of Astronomy Catalogs with Applications to the Hubble Space Telescope

Astrometric calibration of images with a small field of view is often inferior to the internal accuracy of the source detections due to the small number of accessible guide stars. One important experiment with such challenges is the Hubble Space Telescope (HST). A possible solution is to cross-calibrate overlapping fields instead of just relying on standard stars. Following the approach of \citet{2012ApJ...761..188B}, we use infinitesimal 3D rotations for fine-tuning the calibration but devise a better objective that is robust to a large number of false candidates in the initial set of associations. Using Bayesian statistics, we accommodate bad data by explicitly modeling the quality, which yields a formalism essentially identical to an $M$-estimation in robust statistics. Our results on simulated and real catalogs show great potentials for improving the HST calibration, and those with similar challenges.

astro-ph.IM↗

arXiv2017

Photo-z-SQL: integrated, flexible photometric redshift computation in a database

We present a flexible template-based photometric redshift estimation framework, implemented in C#, that can be seamlessly integrated into a SQL database (or DB) server and executed on-demand in SQL. The DB integration eliminates the need to move large photometric datasets outside a database for redshift estimation, and utilizes the computational capabilities of DB hardware. The code is able to perform both maximum likelihood and Bayesian estimation, and can handle inputs of variable photometric filter sets and corresponding broad-band magnitudes. It is possible to take into account the full covariance matrix between filters, and filter zero points can be empirically calibrated using measurements with given redshifts. The list of spectral templates and the prior can be specified flexibly, and the expensive synthetic magnitude computations are done via lazy evaluation, coupled with a caching of results. Parallel execution is fully supported. For large upcoming photometric surveys such as the LSST, the ability to perform in-place photo-z calculation would be a significant advantage. Also, the efficient handling of variable filter sets is a necessity for heterogeneous databases, for example the Hubble Source Catalog, and for cross-match services such as SkyQuery. We illustrate the performance of our code on two reference photo-z estimation testing datasets, and provide an analysis of execution time and scalability with respect to different configurations. The code is available for download at https://github.com/beckrob/Photo-z-SQL.

astro-ph.GA↗

arXiv2016

Mapping the Similarities of Spectra: Global and Locally-biased Approaches to SDSS Galaxy Data

We apply a novel spectral graph technique, that of locally-biased semi-supervised eigenvectors, to study the diversity of galaxies. This technique permits us to characterize empirically the natural variations in observed spectra data, and we illustrate how this approach can be used in an exploratory manner to highlight both large-scale global as well as small-scale local structure in Sloan Digital Sky Survey (SDSS) data. We use this method in a way that simultaneously takes into account the measurements of spectral lines as well as the continuum shape. Unlike Principal Component Analysis, this method does not assume that the Euclidean distance between galaxy spectra is a good global measure of similarity between all spectra, but instead it only assumes that local difference information between similar spectra is reliable. Moreover, unlike other nonlinear dimensionality methods, this method can be used to characterize very finely both small-scale local as well as large-scale global properties of realistic noisy data. The power of the method is demonstrated on the SDSS Main Galaxy Sample by illustrating that the derived embeddings of spectra carry an unprecedented amount of information. By using a straightforward global or unsupervised variant, we observe that the main features correlate strongly with star formation rate and that they clearly separate active galactic nuclei. Computed parameters of the method can be used to describe line strengths and their interdependencies. By using a locally-biased or semi-supervised variant, we are able to focus on typical variations around specific objects of astronomical interest. We present several examples illustrating that this approach can enable new discoveries in the data as well as a detailed understanding of very fine local structure that would otherwise be overwhelmed by large-scale noise and global trends in the data.

astro-ph.IM↗

arXiv2016

Galaxy Redshifts from Discrete Optimization of Correlation Functions

We propose a new method of constraining the redshifts of individual extragalactic sources based on celestial coordinates and their ensemble statistics. Techniques from integer linear programming are utilized to optimize simultaneously for the angular two-point cross- and autocorrelation functions. Our novel formalism introduced here not only transforms the otherwise hopelessly expensive, brute-force combinatorial search into a linear system with integer constraints but also is readily implementable in off-the-shelf solvers. We adopt Gurobi, a commercial optimization solver, and use Python to build the cost function dynamically. The preliminary results on simulated data show potential for future applications to sky surveys by complementing and enhancing photometric redshift estimators. Our approach is the first application of integer linear programming to astronomical analysis.

astro-ph.IM↗

arXiv2016

The Footprint Database and Web Services of the Herschel Space Observatory

Data from the Herschel Space Observatory is freely available to the public but no uniformly processed catalogue of the observations has been published so far. To date, the Herschel Science Archive does not contain the exact sky coverage (footprint) of individual observations and supports search for measurements based on bounding circles only. Drawing on previous experience in implementing footprint databases, we built the Herschel Footprint Database and Web Services for the Herschel Space Observatory to provide efficient search capabilities for typical astronomical queries. The database was designed with the following main goals in mind: (a) provide a unified data model for meta-data of all instruments and observational modes, (b) quickly find observations covering a selected object and its neighbourhood, (c) quickly find every observation in a larger area of the sky, (d) allow for finding solar system objects crossing observation fields. As a first step, we developed a unified data model of observations of all three Herschel instruments for all pointing and instrument modes. Then, using telescope pointing information and observational meta-data, we compiled a database of footprints. As opposed to methods using pixellation of the sphere, we represent sky coverage in an exact geometric form allowing for precise area calculations. For easier handling of Herschel observation footprints with rather complex shapes, two algorithms were implemented to reduce the outline. Furthermore, a new visualisation tool to plot footprints with various spherical projections was developed. Indexing of the footprints using Hierarchical Triangular Mesh makes it possible to quickly find observations based on sky coverage, time and meta-data. The database is accessible via a web site (http://herschel.vo.elte.hu) and also as a set of REST web service functions.

astro-ph.IM↗

arXiv2016

Exploring the SDSS Photometric Galaxies with Clustering Redshifts

We apply clustering-based redshift inference to all extended sources from the Sloan Digital Sky Survey photometric catalogue, down to magnitude r = 22. We map the relationships between colours and redshift, without assumption of the sources' spectral energy distributions (SED). We identify and locate star-forming, quiescent galaxies, and AGN, as well as colour changes due to spectral features, such as the 4000 Å break, redshifting through specific filters. Our mapping is globally in good agreement with colour-redshift tracks computed with SED templates, but reveals informative differences, such as the need for a lower fraction of M-type stars in certain templates. We compare our clustering-redshift estimates to photometric redshifts and find these two independent estimators to be in good agreement at each limiting magnitude considered. Finally, we present the global clustering-redshift distribution of all Sloan extended sources, showing objects up to z ~ 0.8. While the overall shape agrees with that inferred from photometric redshifts, the clustering redshift technique results in a smoother distribution, with no indication of structure in redshift space suggested by the photometric redshift estimates (likely artifacts imprinted by their spectroscopic training set). We also infer a higher fraction of high redshift objects. The mapping between the four observed colours and redshift can be used to estimate the redshift probability distribution function of individual galaxies. This work is an initial step towards producing a general mapping between redshift and all available observables in the photometric space, including brightness, size, concentration, and ellipticity.

astro-ph.GA↗

arXiv2016

Photometric redshifts for the SDSS Data Release 12

We present the methodology and data behind the photometric redshift database of the Sloan Digital Sky Survey Data Release 12 (SDSS DR12). We adopt a hybrid technique, empirically estimating the redshift via local regression on a spectroscopic training set, then fitting a spectrum template to obtain K-corrections and absolute magnitudes. The SDSS spectroscopic catalog was augmented with data from other, publicly available spectroscopic surveys to mitigate target selection effects. The training set is comprised of $1,976,978$ galaxies, and extends up to redshift $z\approx 0.8$, with a useful coverage of up to $z\approx 0.6$. We provide photometric redshifts and realistic error estimates for the $208,474,076$ galaxies of the SDSS primary photometric catalog. We achieve an average bias of $\overline{Δz_{\mathrm{norm}}} = 5.84 \times 10^{-5}$, a standard deviation of $σ\left(Δz_{\mathrm{norm}}\right)=0.0205$, and a $3σ$ outlier rate of $P_o=4.11\%$ when cross-validating on our training set. The published redshift error estimates and photometric error classes enable the selection of galaxies with high quality photometric redshifts. We also provide a supplementary error map that allows additional, sophisticated filtering of the data.

astro-ph.GA↗

arXiv2015

Matching Radio Catalogs with Realistic Geometry: Application to SWIRE and ATLAS

Crossmatching catalogs at different wavelengths is a difficult problem in astronomy, especially when the objects are not point-like. At radio wavelengths an object can have several components corresponding, for example, to a core and lobes. {Considering not all radio detections correspond to visible or infrared sources, matching these catalogs can be challenging.} Traditionally this is done by eye for better quality, which does not scale to the large data volumes expected from the next-generation of radio telescopes. We present a novel automated procedure, using Bayesian hypothesis testing, to achieve reliable associations by explicit modelling of a particular class of radio-source morphology. {The new algorithm not only assesses the likelihood of an association between data at two different wavelengths, but also tries to assess whether different radio sources are physically associated, are double-lobed radio galaxies, or just distinct nearby objects.} Application to the SWIRE and ATLAS CDF-S catalogs shows that this method performs well without human intervention.

astro-ph.IM↗