arXiv subjects

Y. Shu

Publications and source records attributed to Y. Shu.

14 recordsLinked to original sources

arXiv2025

DESI Strong Lens Foundry II: DESI Spectroscopy for Strong Lens Candidates

We present the Dark Energy Spectroscopic Instrument (DESI) Strong Lensing Secondary Target Program. This is a spectroscopic follow-up program for strong gravitational lens candidates found in the DESI Legacy Imaging Surveys footprint. Spectroscopic redshifts for the lenses and lensed source are crucial for lens modeling to obtain physical parameters. The spectroscopic catalog in this paper consists of 73 candidate systems from the DESI Early Data Release (EDR). We have confirmed 20 strong lensing systems and determined four to not be lenses. For the remaining systems, more spectroscopic data from ongoing and future observations will be presented in future publications. We discuss the implications of our results for lens searches with neural networks in existing and future imaging surveys as well as for lens modeling. This Strong Lensing Secondary Target Program is part of the DESI Strong Lens Foundry project, and this is Paper II of a series on this project.

astro-ph.CO↗

arXiv2025

DESI Strong Lens Foundry III: Keck Spectroscopy for Strong Lenses Discovered Using Residual Neural Networks

We present spectroscopic data of strong lenses and their source galaxies using the Keck Near-Infrared Echellette Spectrometer (NIRES) and the Dark Energy Spectroscopic Instrument (DESI), providing redshifts necessary for nearly all strong-lensing applications with these systems, especially the extraction of physical parameters from lensing modeling. These strong lenses were found in the DESI Legacy Imaging Surveys using Residual Neural Networks (ResNet) and followed up by our Hubble Space Telescope program, with all systems displaying unambiguous lensed arcs. With NIRES, we target eight lensed sources at redshifts difficult to measure in the optical range and determine the source redshifts for six, between $z_s$ = 1.675 and 3.332. DESI observed one of the remaining source redshifts, as well as an additional source redshift within the six systems. The two systems with non-detections by NIRES were observed for a considerably shorter 600s at high airmass. Combining NIRES infrared spectroscopy with optical spectroscopy from our DESI Strong Lensing Secondary Target Program, these results provide the complete lens and source redshifts for six systems, a resource for refining automated strong lens searches in future deep- and wide-field imaging surveys and addressing a range of questions in astrophysics and cosmology.

astro-ph.CO↗

arXiv2025

HOLISMOKES XVI: Lens search in HSC-PDR3 with a neural network committee and post-processing for false-positive removal

We have carried out a systematic search for galaxy-scale lenses exploiting multi-band imaging data from the third public data release of the Hyper Suprime-Cam (HSC) survey with the focus on false-positive removal, after applying deep learning classifiers to all 110 million sources with i-Kron radius above 0.8". To improve the performance, we tested the combination of multiple networks from our previous lens search projects and found the best performance by averaging the scores from five of our networks. Although this ensemble network leads already to a false-positive rate (FPR) of 0.01% at a true-positive rate (TPR) of 75% on known real lenses, we have elaborated techniques to further clean the network candidate list before visual inspection. In detail, we tested the rejection using SExtractor and the modeling network from HOLISMOKES IX, which resulted together in a candidate rejection of 29% without lowering the TPR. We carried out a comprehensive multi-stage visual inspection involving eight individuals and identified 95 grade A (average grade G >2.5) and 503 grade B (2.5 >G >1.5) lens candidates, including 92 discoveries reported for the first time. This inspection also incorporated a novel environmental characterization using histograms of photometric redshifts. We publicly release the average grades, mass model predictions, and environment characterization of all visually inspected candidates, while including references for previously discovered systems, which makes this catalog one of the largest compilation of known lenses. The results demonstrate that (1) the combination of multiple networks enhances the selection performance and (2) both automated masking tools as well as modeling networks, which can be easily applied to hundreds of thousands of network candidates, help reduce the number of false positives that is the main limitation in lens search to date.

astro-ph.IM↗

arXiv2025

DESI Strong Lens Foundry I: HST Observations and Modeling with GIGA-Lens

We present the Dark Energy Spectroscopic Instrument (DESI) Strong Lens Foundry. We discovered $\sim 3500$ new strong gravitational lens candidates in the DESI Legacy Imaging Surveys using residual neural networks (ResNet). We observed a subset (51) of our candidates using the Hubble Space Telescope (HST). All of them were confirmed to be strong lenses. We also briefly describe spectroscopic follow-up observations by DESI and Keck NIRES programs. From this very rich dataset, a number of studies will be carried out, including evaluating the quality of the ResNet search candidates and lens modeling. In this paper, we present our initial effort in these directions. In particular, as a demonstration, we present the lens model for DESI-165.4754-06.0423, with imaging data from HST, and lens and source redshifts from DESI and Keck NIRES, respectively. In this effort, we have applied a \emph{fully} forward-modeling Bayesian approach (GIGA-Lens), using \emph{multiple} GPUs, for the first time in both regards, to a strong lens with HST data, or any high resolution imaging.

astro-ph.CO↗

arXiv2023

HOLISMOKES -- XI. Evaluation of supervised neural networks for strong-lens searches in ground-based imaging surveys

While supervised neural networks have become state of the art for identifying the rare strong gravitational lenses from large imaging data sets, their selection remains significantly affected by the large number and diversity of nonlens contaminants. This work evaluates and compares systematically the performance of neural networks in order to move towards a rapid selection of galaxy-scale strong lenses with minimal human input in the era of deep, wide-scale surveys. We used multiband images from PDR2 of the HSC Wide survey to build test sets mimicking an actual classification experiment, with 189 strong lenses previously found over the HSC footprint and 70,910 nonlens galaxies in COSMOS. Multiple networks were trained on different sets of realistic strong-lens simulations and nonlens galaxies, with various architectures and data pre-processing. The overall performances strongly depend on the construction of the ground-truth training data and they typically, but not systematically, improve using our baseline residual network architecture. Improvements are found when applying random shifts to the image centroids and square root stretches to the pixel values, adding z band, or using random viewpoints of the original images, but not when adding difference images to subtract emission from the central galaxy. The most significant gain is obtained with committees of networks trained on different data sets, and showing a moderate overlap between populations of false positives. Nearly-perfect invariance to image quality can be achieved by training networks either with large number of bands, or jointly with the PSF and science frames. Overall, we show the possibility to reach a TPR0 as high as 60% for the test sets under consideration, which opens promising perspectives for pure selection of strong lenses without human input using the Rubin Observatory and other forthcoming ground-based surveys.

astro-ph.GA↗

arXiv2022

LensWatch: I. Resolved HST Observations and Constraints on the Strongly-Lensed Type Ia Supernova 2022qmx ("SN Zwicky")

Supernovae (SNe) that have been multiply-imaged by gravitational lensing are rare and powerful probes for cosmology. Each detection is an opportunity to develop the critical tools and methodologies needed as the sample of lensed SNe increases by orders of magnitude with the upcoming Vera C. Rubin Observatory and Nancy Grace Roman Space Telescope. The latest such discovery is of the quadruply-imaged Type Ia SN 2022qmx (aka, "SN Zwicky") at $z=0.3544$. SN Zwicky was discovered by the Zwicky Transient Facility (ZTF) in spatially unresolved data. Here we present follow-up Hubble Space Telescope observations of SN Zwicky, the first from the multi-cycle "LensWatch" program. We measure photometry for each of the four images of SN Zwicky, which are resolved in three WFC3/UVIS filters (F475W, F625W, F814W) but unresolved with WFC3/IR~F160W, and present an analysis of the lensing system using a variety of independent lens modeling methods. We find consistency between lens model predicted time delays ($\lesssim1$ day), and delays estimated with the single epoch of HST colors ($\lesssim3.5$ days), including the uncertainty from chromatic microlensing ($\sim1$-$1.5$ days). Our lens models converge to an Einstein radius of $\theta_E=(0.168^{+0.009}_{-0.005})\prime\prime$, the smallest yet seen in a lensed SN system. The "standard candle" nature of SN Zwicky provides magnification estimates independent of the lens modeling that are brighter than predicted by $\sim1.7^{+0.8}_{-0.6}$mag and $\sim0.9^{+0.8}_{-0.6}$mag for two of the four images, suggesting significant microlensing and/or additional substructure beyond the flexibility of our image-position mass models.

astro-ph.CO↗

arXiv2022

HOLISMOKES -- X. Comparison between neural network and semi-automated traditional modeling of strong lenses

Modeling of strongly gravitationally lensed galaxies is often required in order to use them as astrophysical or cosmological probes. With current and upcoming wide-field imaging surveys, the number of detected lenses is increasing significantly such that automated and fast modeling procedures for ground-based data are urgently needed. This is especially pertinent to short-lived lensed transients in order to plan follow-up observations. Therefore, we present in a companion paper (submitted) a neural network predicting the parameter values with corresponding uncertainties of a Singular Isothermal Ellipsoid (SIE) mass profile with external shear. In this work, we present a newly-developed pipeline glee_auto.py to model consistently any galaxy-scale lensing system. In contrast to previous automated modeling pipelines that require high-resolution images, glee_auto.py is optimized for ground-based images such as those from the Hyper-Suprime-Cam (HSC) or the upcoming Rubin Observatory Legacy Survey of Space and Time. We further present glee_tools.py, a flexible automation code for individual modeling that has no direct decisions and assumptions implemented. Both pipelines, in addition to our modeling network, minimize the user input time drastically and thus are important for future modeling efforts. We apply the network to 31 real galaxy-scale lenses of HSC and compare the results to the traditional models. In the direct comparison, we find a very good match for the Einstein radius especially for systems with $\theta_E \gtrsim 2$". The lens mass center and ellipticity show reasonable agreement. The main discrepancies are on the external shear as expected from our tests on mock systems. In general, our study demonstrates that neural networks are a viable and ultra fast approach for measuring the lens-galaxy masses from ground-based data in the upcoming era with $\sim10^5$ lenses expected.

astro-ph.CO↗

arXiv2022

HOLISMOKES -- IX. Neural network inference of strong-lens parameters and uncertainties from ground-based images

Modeling of strong gravitational lenses is a necessity for further applications in astrophysics and cosmology. Especially with the large number of detections in current and upcoming surveys such as the Rubin Legacy Survey of Space and Time (LSST), it is timely to investigate in automated and fast analysis techniques beyond the traditional and time consuming Markov chain Monte Carlo sampling methods. Building upon our convolutional neural network (CNN) presented in Schuldt et al. (2021b), we present here another CNN, specifically a residual neural network (ResNet), that predicts the five mass parameters of a Singular Isothermal Ellipsoid (SIE) profile (lens center $x$ and $y$, ellipticity $e_x$ and $e_y$, Einstein radius $\theta_E$) and the external shear ($\gamma_{ext,1}$, $\gamma_{ext,2}$) from ground-based imaging data. In contrast to our CNN, this ResNet further predicts a 1$\sigma$ uncertainty for each parameter. To train our network, we use our improved pipeline from Schuldt et al. (2021b) to simulate lens images using real images of galaxies from the Hyper Suprime-Cam Survey (HSC) and from the Hubble Ultra Deep Field as lens galaxies and background sources, respectively. We find overall very good recoveries for the SIE parameters, while differences remain in predicting the external shear. From our tests, most likely the low image resolution is the limiting factor for predicting the external shear. Given the run time of milli-seconds per system, our network is perfectly suited to predict the next appearing image and time delays of lensed transients in time. Therefore, we also present the performance of the network on these quantities in comparison to our simulations. Our ResNet is able to predict the SIE and shear parameter values in fractions of a second on a single CPU such that we are able to process efficiently the huge amount of expected galaxy-scale lenses in the near future.

astro-ph.IM↗

arXiv2022

GIGA-Lens: Fast Bayesian Inference for Strong Gravitational Lens Modeling

We present GIGA-Lens: a gradient-informed, GPU-accelerated Bayesian framework for modeling strong gravitational lensing systems, implemented in TensorFlow and JAX. The three components, optimization using multi-start gradient descent, posterior covariance estimation with variational inference, and sampling via Hamiltonian Monte Carlo, all take advantage of gradient information through automatic differentiation and massive parallelization on graphics processing units (GPUs). We test our pipeline on a large set of simulated systems and demonstrate in detail its high level of performance. The average time to model a single system on four Nvidia A100 GPUs is 105 seconds. The robustness, speed, and scalability offered by this framework make it possible to model the large number of strong lenses found in current surveys and present a very promising prospect for the modeling of $\mathcal{O}(10^5)$ lensing systems expected to be discovered in the era of the Vera C. Rubin Observatory, Euclid, and the Nancy Grace Roman Space Telescope.

astro-ph.IM↗

arXiv2021

HOLISMOKES. VI. New galaxy-scale strong lens candidates from the HSC-SSP imaging survey

We have carried out a systematic search for galaxy-scale strong lenses in multiband imaging from the Hyper Suprime-Cam (HSC) survey. Our automated pipeline, based on realistic strong-lens simulations, deep neural network classification, and visual inspection, is aimed at efficiently selecting systems with wide image separations (Einstein radii ~1.0-3.0"), intermediate redshift lenses (z ~ 0.4-0.7), and bright arcs for galaxy evolution and cosmology. We classified gri images of all 62.5 million galaxies in HSC Wide with i-band Kron radius >0.8" to avoid strict pre-selections and to prepare for the upcoming era of deep, wide-scale imaging surveys with Euclid and Rubin Observatory. We obtained 206 newly-discovered candidates classified as definite or probable lenses with either spatially-resolved multiple images or extended, distorted arcs. In addition, we found 88 high-quality candidates that were assigned lower confidence in previous HSC searches, and we recovered 173 known systems in the literature. These results demonstrate that, aided by limited human input, deep learning pipelines with false positive rates as low as ~0.01% can be very powerful tools for identifying the rare strong lenses from large catalogs, and can also largely extend the samples found by traditional algorithms. We provide a ranked list of candidates for future spectroscopic confirmation.

astro-ph.GA↗

arXiv2020

The discovery of the most UV-Lya luminous star-forming galaxy: a young, dust- and metal-poor starburst with QSO-like luminosities

We report the discovery of BOSS-EUVLG1 at z=2.469, by far the most luminous, almost un-obscured star-forming galaxy known at any redshift. First classified as a QSO within the Baryon Oscillation Spectroscopic Survey, follow-up observations with the Gran Telescopio Canarias reveal that its large luminosity, MUV = -24.40 and log(L_Lya/erg s-1) = 44.0, is due to an intense burst of star-formation, and not to an AGN or gravitational lensing. BOSS-EUVLG1 is a compact (reff = 1.2 kpc), young (4-5 Myr) starburst with a stellar mass log(M*/Msun) = 10.0 +/- 0.1 and a prodigious star formation rate of ~1000 Msun yr-1. However, it is metal- and dust-poor (12+log(O/H) = 8.13 +/- 0.19, E(B-V) = 0.07, log(LIR/LUV) < -1.2), indicating that we are witnessing the very early phase of an intense starburst that has had no time to enrich the ISM. BOSS-EUVLG1 might represent a short-lived (<100 Myrs), yet important phase of star-forming galaxies at high redshift that has been missed in previous surveys. Within a galaxy evolutionary scheme, BOSS-EUVLG1 could likely represent the very initial phases in the evolution of massive quiescent galaxies, even before the dusty star-forming phase.

astro-ph.GA↗

arXiv2019

Rest-frame UV Properties of Luminous Strong Gravitationally Lensed Ly$α$ Emitters from the BELLS GALLERY Survey

We present deep rest-frame UV spectroscopic observations using the Gran Telescopio Canarias of six gravitationally lensed Lya emitters (LAEs) at $2.36<z<2.82$ selected from the BELLS GALLERY survey. By taking the magnifications into account, we show that LAEs can be as luminous as L(Lya) = 30x10$^{42}$ erg s-1 and M(UV) = -23 (AB) without invoking an AGN component, in contrast with previous findings. We measure Lya rest-frame equivalent widths, EW(Lya), ranging from 16Åto 50Åand Lya escape fractions, fesc(Lya), from 10% to 40%. Large EW(Lya) and fesc(Lya) are found predominantly in LAEs showing weak low-ionization ISM absorption (EW < 1Å) and narrow Lya profiles (< 300 km s-1 FWHM) with their peak close (< 80 km s-1) to their systemic redshifts, suggestive of less scatter from low HI column densities that favours the escape of Lya photons. We infer stellar metallicities of Z/Zsun ~ 0.2 in almost all LAEs by comparing the P-Cygni profiles of the wind lines NV1240Åand CIV1549Åwith those from stellar synthesis models. We also find a trend between M(UV) and the velocity offset of ISM absorption lines, such as the most luminous LAEs experience stronger outflows. The most luminous LAEs show star formation rates up to 180 Msun yr-1, yet they appear relatively blue ($β$(UV) ~ -1.8 to -2.0) showing evidence of little dust attenuation (E(B-V) = 0.10-0.14). These luminous LAEs may be particular cases of young starburst galaxies that have had no time to form large amounts of dust. If so, they are ideal laboratories to study the early phase of massive star formation, stellar and dust mass growth, and chemical enrichment histories of starburst galaxies at high-z.

astro-ph.GA↗

arXiv2019

Discovery of a giant and luminous Lya+CIV+HeII nebula at z=3.326 with extreme emission line ratios

We present the discovery of HLock01-LAB, a luminous and large Lya nebula at z=3.326. Medium-band imaging and long-slit spectroscopic observations with the Gran Telescopio Canarias reveal extended emission in the Lya 1215Å, CIV1550Å, and HeII 1640Ålines over ~100kpc, and a total luminosity L(Lya)=(6.4+/-0.1)x10^44 erg s^-1. HLock01-LAB presents an elongated morphology aligned with two faint radio sources contained within the central ~8kpc of the nebula. The radio structures are consistent to be faint radio jets or lobes of a central galaxy, whose spectrum shows nebular emission characteristic of a type-II active galactic nucleus (AGN). The continuum emission of the AGN at short wavelengths is, however, likely dominated by stellar emission of the host galaxy, for which we derive a stellar mass M* = 2.3x10^11 Msun. The detection of extended emission in CIV and CIII] indicates that the gas within the nebula is not primordial. Feedback may have enriched the halo at at least 50 kpc from the nuclear region. Using rest-frame UV emission-line diagnostics, we find that the gas in the nebula is likely heated by the AGN. Nevertheless, at the center of the nebula we find extreme emission line ratios of Lya/CIV~60 and Lya/HeII~80, one of the highest values measured to date, and well above the standard values of photoionization models (Lya/HeII~30 for case B photoionization). Our data suggest that jet-induced shocks are likely responsible for the increase of the electron temperature and, thus, the observed Lya enhancement in the center of the nebula. This scenario is further supported by the presence of radio structures and perturbed kinematics in this region. The large Lya luminosity in HLock01-LAB is likely due to a combination of AGN photoionization and jet-induced shocks, highlighting the diversity of sources of energy powering Lya nebulae. [abridged]

astro-ph.GA↗

arXiv2019

Chinese Sunspot Drawing and Its Digitization-(I) Parameter Archives

Based on the Chinese historical sunspots drawings, a data set consisting of the scanned images and all their digitized parameters from 1925 to 2015 have been constructed. In this paper, we briefly describe the developmental history of sunspots drawings in China. This paper describes the preliminary processing processes that strat from the initial data (inputing to the scanning equipment) to the parameters extraction, and finally summarizes the general features of this dataset. It is the first systematic project in Chinese solar-physics community that the historical observation of sunspots drawings were digitized. Our data set fills in an almost ninety years historical gap, which span 60 degrees from east to west and 50 degrees from north to south and have no continuous and detailed digital sunspot observation information. As a complementary to other sunspots observation in the world, our dataset provided abundant information to the long term solar cycles solar activity research.

astro-ph.SR↗