arXiv subjects

B. Nord

Publications and source records attributed to B. Nord.

At least 19 recordsLinked to original sources

arXiv2025

Discovering Strong Gravitational Lenses in the Dark Energy Survey with Interactive Machine Learning and Crowd-sourced Inspection with Space Warps

We conduct a search for strong gravitational lenses in the Dark Energy Survey (DES) Year 6 imaging data. We implement a pre-trained Vision Transformer (ViT) for our machine learning (ML) architecture and adopt Interactive Machine Learning to construct a training sample with multiple classes to address common types of false positives. Our ML model reduces 236 million DES cutout images to 22,564 targets of interest, including around 85% of previously reported galaxy-galaxy lens candidates discovered in DES. These targets were visually inspected by citizen scientists, who ruled out approximately 90% as false positives. Of the remaining 2,618 candidates, 149 were expert-classified as 'definite' lenses and 516 as 'probable' lenses, with 147 of these candidates being newly identified. Additionally, we trained a second ViT to find double-source plane lens systems, finding at least one double-source system. Our main ViT excels at identifying galaxy-galaxy lenses, consistently assigning high scores to candidates with high confidence. The top 800 ViT-scored images include around 100 of our `definite' lens candidates. This selection is an order of magnitude higher in purity than previous convolutional neural network-based lens searches and demonstrates the feasibility of applying our methodology for discovering large samples of lenses in future surveys.

astro-ph.GA↗

arXiv2023

Constructing Impactful Machine Learning Research for Astronomy: Best Practices for Researchers and Reviewers

Machine learning has rapidly become a tool of choice for the astronomical community. It is being applied across a wide range of wavelengths and problems, from the classification of transients to neural network emulators of cosmological simulations, and is shifting paradigms about how we generate and report scientific results. At the same time, this class of method comes with its own set of best practices, challenges, and drawbacks, which, at present, are often reported on incompletely in the astrophysical literature. With this paper, we aim to provide a primer to the astronomical community, including authors, reviewers, and editors, on how to implement machine learning models and report their results in a way that ensures the accuracy of the results, reproducibility of the findings, and usefulness of the method.

astro-ph.IM↗

arXiv2023

DeepAstroUDA: Semi-Supervised Universal Domain Adaptation for Cross-Survey Galaxy Morphology Classification and Anomaly Detection

Artificial intelligence methods show great promise in increasing the quality and speed of work with large astronomical datasets, but the high complexity of these methods leads to the extraction of dataset-specific, non-robust features. Therefore, such methods do not generalize well across multiple datasets. We present a universal domain adaptation method, \textit{DeepAstroUDA}, as an approach to overcome this challenge. This algorithm performs semi-supervised domain adaptation and can be applied to datasets with different data distributions and class overlaps. Non-overlapping classes can be present in any of the two datasets (the labeled source domain, or the unlabeled target domain), and the method can even be used in the presence of unknown classes. We apply our method to three examples of galaxy morphology classification tasks of different complexities ($3$-class and $10$-class problems), with anomaly detection: 1) datasets created after different numbers of observing years from a single survey (LSST mock data of $1$ and $10$ years of observations); 2) data from different surveys (SDSS and DECaLS); and 3) data from observing fields with different depths within one survey (wide field and Stripe 82 deep field of SDSS). For the first time, we demonstrate the successful use of domain adaptation between very discrepant observational datasets. \textit{DeepAstroUDA} is capable of bridging the gap between two astronomical surveys, increasing classification accuracy in both domains (up to $40\%$ on the unlabeled data), and making model performance consistent across datasets. Furthermore, our method also performs well as an anomaly detection algorithm and successfully clusters unknown class samples even in the unlabeled target dataset.

astro-ph.GA↗

arXiv2022

DeepZipper II: Searching for Lensed Supernovae in Dark Energy Survey Data with Deep Learning

Gravitationally lensed supernovae (LSNe) are important probes of cosmic expansion, but they remain rare and difficult to find. Current cosmic surveys likely contain and 5-10 LSNe in total while next-generation experiments are expected to contain several hundreds to a few thousands of these systems. We search for these systems in observed Dark Energy Survey (DES) 5-year SN fields -- 10 3-sq. deg. regions of sky imaged in the $griz$ bands approximately every six nights over five years. To perform the search, we utilize the DeepZipper approach: a multi-branch deep learning architecture trained on image-level simulations of LSNe that simultaneously learns spatial and temporal relationships from time series of images. We find that our method obtains a LSN recall of 61.13% and a false positive rate of 0.02% on the DES SN field data. DeepZipper selected 2,245 candidates from a magnitude-limited ($m_i$ $<$ 22.5) catalog of 3,459,186 systems. We employ human visual inspection to review systems selected by the network and find three candidate LSNe in the DES SN fields.

astro-ph.CO↗

arXiv2022

DeepZipper: A Novel Deep Learning Architecture for Lensed Supernovae Identification

Large-scale astronomical surveys have the potential to capture data on large numbers of strongly gravitationally lensed supernovae (LSNe). To facilitate timely analysis and spectroscopic follow-up before the supernova fades, an LSN needs to be identified soon after it begins. To quickly identify LSNe in optical survey datasets, we designed ZipperNet, a multi-branch deep neural network that combines convolutional layers (traditionally used for images) with long short-term memory (LSTM) layers (traditionally used for time series). We tested ZipperNet on the task of classifying objects from four categories -- no lens, galaxy-galaxy lens, lensed type Ia supernova, lensed core-collapse supernova -- within high-fidelity simulations of three cosmic survey data sets -- the Dark Energy Survey (DES), Rubin Observatory's Legacy Survey of Space and Time (LSST), and a Dark Energy Spectroscopic Instrument (DESI) imaging survey. Among our results, we find that for the LSST-like dataset, ZipperNet classifies LSNe with a receiver operating characteristic area under the curve of 0.97, predicts the spectroscopic type of the lensed supernovae with 79\% accuracy, and demonstrates similarly high performance for LSNe 1-2 epochs after first detection. We anticipate that a model like ZipperNet, which simultaneously incorporates spatial and temporal information, can play a significant role in the rapid identification of lensed transient systems in cosmic survey experiments.

astro-ph.CO↗

arXiv2022

The DES Bright Arcs Survey: Candidate Strongly Lensed Galaxy Systems from the Dark Energy Survey 5,000 Sq. Deg. Footprint

We report the combined results of eight searches for strong gravitational lens systems in the full 5,000 sq. deg. of Dark Energy Survey (DES) observations. The observations accumulated by the end of the third observing season fully covered the DES footprint in 5 filters (grizY), with an $i-$band limiting magnitude (at $10σ$) of 23.44. In four searches, a list of potential candidates was identified using a color and magnitude selection from the object catalogs created from the first three observing seasons. Three other searches were conducted at the locations of previously identified galaxy clusters. Cutout images of potential candidates were then visually scanned using an object viewer. An additional set of candidates came from a data-quality check of a subset of the color-coadd "tiles" created from the full DES six-season data set. A short list of the most promising strong lens candidates was then numerically ranked according to whether or not we judged them to be bona fide strong gravitational lens systems. These searches discovered a diverse set of 247 strong lens candidate systems, of which 81 are identified for the first time. We provide the coordinates, magnitudes, and photometric properties of the lens and source objects, and an estimate of the Einstein radius for 81 new systems and 166 previously reported. This catalog will be of use for selecting interesting systems for detailed follow-up, studies of galaxy cluster and group mass profiles, as well as a training/validation set for automated strong lens searches.

astro-ph.CO↗

arXiv2021

Robustness of deep learning algorithms in astronomy -- galaxy morphology studies

Deep learning models are being increasingly adopted in wide array of scientific domains, especially to handle high-dimensionality and volume of the scientific data. However, these models tend to be brittle due to their complexity and overparametrization, especially to the inadvertent adversarial perturbations that can appear due to common image processing such as compression or blurring that are often seen with real scientific data. It is crucial to understand this brittleness and develop models robust to these adversarial perturbations. To this end, we study the effect of observational noise from the exposure time, as well as the worst case scenario of a one-pixel attack as a proxy for compression or telescope errors on performance of ResNet18 trained to distinguish between galaxies of different morphologies in LSST mock data. We also explore how domain adaptation techniques can help improve model robustness in case of this type of naturally occurring attacks and help scientists build more trustworthy and stable models.

astro-ph.GA↗

arXiv2021

Finding quadruply imaged quasars with machine learning. I. Methods

Strongly lensed quadruply imaged quasars (quads) are extraordinary objects. They are very rare in the sky -- only a few tens are known to date -- and yet they provide unique information about a wide range of topics, including the expansion history and the composition of the Universe, the distribution of stars and dark matter in galaxies, the host galaxies of quasars, and the stellar initial mass function. Finding them in astronomical images is a classic "needle in a haystack" problem, as they are outnumbered by other (contaminant) sources by many orders of magnitude. To solve this problem, we develop state-of-the-art deep learning methods and train them on realistic simulated quads based on real images of galaxies taken from the Dark Energy Survey, with realistic source and deflector models, including the chromatic effects of microlensing. The performance of the best methods on a mixture of simulated and real objects is excellent, yielding area under the receiver operating curve in the range 0.86 to 0.89. Recall is close to 100% down to total magnitude i~21 indicating high completeness, while precision declines from 85% to 70% in the range i~17-21. The methods are extremely fast: training on 2 million samples takes 20 hours on a GPU machine, and 10^8 multi-band cutouts can be evaluated per GPU-hour. The speed and performance of the method pave the way to apply it to large samples of astronomical sources, bypassing the need for photometric pre-selection that is likely to be a major cause of incompleteness in current samples of known quads.

astro-ph.CO↗

arXiv2021

Density split statistics: joint model of counts and lensing in cells

We present density split statistics, a framework that studies lensing and counts-in-cells as a function of foreground galaxy density, thereby providing a large-scale measurement of both 2-point and 3-point statistics. Our method extends our earlier work on trough lensing and is summarized as follows: given a foreground (low redshift) population of galaxies, we divide the sky into subareas of equal size but distinct galaxy density. We then measure lensing around uniformly spaced points separately in each of these subareas, as well as counts-in-cells statistics (CiC). The lensing signals trace the matter density contrast around regions of fixed galaxy density. Through the CiC measurements this can be related to the density profile around regions of fixed matter density. Together, these measurements constitute a powerful probe of cosmology, the skewness of the density field and the connection of galaxies and matter. In this paper we show how to model both the density split lensing signal and CiC from basic ingredients: a non-linear power spectrum, clustering hierarchy coefficients from perturbation theory and a parametric model for galaxy bias and shot-noise. Using N-body simulations, we demonstrate that this model is sufficiently accurate for a cosmological analysis on year 1 data from the Dark Energy Survey.

astro-ph.CO↗

arXiv2021

DeepMerge II: Building Robust Deep Learning Algorithms for Merging Galaxy Identification Across Domains

In astronomy, neural networks are often trained on simulation data with the prospect of being used on telescope observations. Unfortunately, training a model on simulation data and then applying it to instrument data leads to a substantial and potentially even detrimental decrease in model accuracy on the new target dataset. Simulated and instrument data represent different data domains, and for an algorithm to work in both, domain-invariant learning is necessary. Here we employ domain adaptation techniques$-$ Maximum Mean Discrepancy (MMD) as an additional transfer loss and Domain Adversarial Neural Networks (DANNs)$-$ and demonstrate their viability to extract domain-invariant features within the astronomical context of classifying merging and non-merging galaxies. Additionally, we explore the use of Fisher loss and entropy minimization to enforce better in-domain class discriminability. We show that the addition of each domain adaptation technique improves the performance of a classifier when compared to conventional deep learning algorithms. We demonstrate this on two examples: between two Illustris-1 simulated datasets of distant merging galaxies, and between Illustris-1 simulated data of nearby merging galaxies and observed data from the Sloan Digital Sky Survey. The use of domain adaptation techniques in our experiments leads to an increase of target domain classification accuracy of up to ${\sim}20\%$. With further development, these techniques will allow astronomers to successfully implement neural network models trained on simulation data to efficiently detect and study astrophysical objects in current and future large-scale astronomical surveys.

astro-ph.IM↗

arXiv2020

Domain adaptation techniques for improved cross-domain study of galaxy mergers

In astronomy, neural networks are often trained on simulated data with the prospect of being applied to real observations. Unfortunately, simply training a deep neural network on images from one domain does not guarantee satisfactory performance on new images from a different domain. The ability to share cross-domain knowledge is the main advantage of modern deep domain adaptation techniques. Here we demonstrate the use of two techniques - Maximum Mean Discrepancy (MMD) and adversarial training with Domain Adversarial Neural Networks (DANN) - for the classification of distant galaxy mergers from the Illustris-1 simulation, where the two domains presented differ only due to inclusion of observational noise. We show how the addition of either MMD or adversarial training greatly improves the performance of the classifier on the target domain when compared to conventional machine learning algorithms, thereby demonstrating great promise for their use in astronomy.

astro-ph.IM↗

arXiv2020

DeepMerge: Classifying High-redshift Merging Galaxies with Deep Neural Networks

We investigate and demonstrate the use of convolutional neural networks (CNNs) for the task of distinguishing between merging and non-merging galaxies in simulated images, and for the first time at high redshifts (i.e. $z=2$). We extract images of merging and non-merging galaxies from the Illustris-1 cosmological simulation and apply observational and experimental noise that mimics that from the Hubble Space Telescope; the data without noise form a "pristine" data set and that with noise form a "noisy" data set. The test set classification accuracy of the CNN is $79\%$ for pristine and $76\%$ for noisy. The CNN outperforms a Random Forest classifier, which was shown to be superior to conventional one- or two-dimensional statistical methods (Concentration, Asymmetry, the Gini, $M_{20}$ statistics etc.), which are commonly used when classifying merging galaxies. We also investigate the selection effects of the classifier with respect to merger state and star formation rate, finding no bias. Finally, we extract Grad-CAMs (Gradient-weighted Class Activation Mapping) from the results to further assess and interrogate the fidelity of the classification model.

astro-ph.GA↗

arXiv2019

Petabytes to Science

A Kavli foundation sponsored workshop on the theme \emph{Petabytes to Science} was held 12$^{th}$ to 14$^{th}$ of February 2019 in Las Vegas. The aim of the this workshop was to discuss important trends and technologies which may support astronomy. We also tackled how to better shape the workforce for the new trends and how we should approach education and public outreach. This document was coauthored during the workshop and edited in the weeks after. It comprises the discussions and highlights many recommendations which came out of the workshop. We shall distill parts of this document and formulate potential white papers for the decadal survey.

astro-ph.IM↗

arXiv2019

Dark Energy Survey Year 1 Results: Constraints on Extended Cosmological Models from Galaxy Clustering and Weak Lensing

We present constraints on extensions of the minimal cosmological models dominated by dark matter and dark energy, $Λ$CDM and $w$CDM, by using a combined analysis of galaxy clustering and weak gravitational lensing from the first-year data of the Dark Energy Survey (DES Y1) in combination with external data. We consider four extensions of the minimal dark energy-dominated scenarios: 1) nonzero curvature $Ω_k$, 2) number of relativistic species $N_{\rm eff}$ different from the standard value of 3.046, 3) time-varying equation-of-state of dark energy described by the parameters $w_0$ and $w_a$ (alternatively quoted by the values at the pivot redshift, $w_p$, and $w_a$), and 4) modified gravity described by the parameters $μ_0$ and $Σ_0$ that modify the metric potentials. We also consider external information from Planck CMB measurements; BAO measurements from SDSS, 6dF, and BOSS; RSD measurements from BOSS; and SNIa information from the Pantheon compilation. Constraints on curvature and the number of relativistic species are dominated by the external data; when these are combined with DES Y1, we find $Ω_k=0.0020^{+0.0037}_{-0.0032}$ at the 68% confidence level, and $N_{\rm eff}<3.28\, (3.55)$ at 68% (95%) confidence. For the time-varying equation-of-state, we find the pivot value $(w_p, w_a)=(-0.91^{+0.19}_{-0.23}, -0.57^{+0.93}_{-1.11})$ at pivot redshift $z_p=0.27$ from DES alone, and $(w_p, w_a)=(-1.01^{+0.04}_{-0.04}, -0.28^{+0.37}_{-0.48})$ at $z_p=0.20$ from DES Y1 combined with external data; in either case we find no evidence for the temporal variation of the equation of state. For modified gravity, we find the present-day value of the relevant parameters to be $Σ_0= 0.43^{+0.28}_{-0.29}$ from DES Y1 alone, and $(Σ_0, μ_0)=(0.06^{+0.08}_{-0.07}, -0.11^{+0.42}_{-0.46})$ from DES Y1 combined with external data, consistent with predictions from GR.

astro-ph.CO↗

arXiv2019

Observation and Confirmation of Nine Strong Lensing Systems in Dark Energy Survey Year 1 Data

We describe the observation and confirmation of \nbconfirmtext\ new strong gravitational lenses discovered in Year 1 data from the Dark Energy Survey (DES). We created candidate lists based on a) galaxy group and cluster samples and b) photometrically selected galaxy samples. We selected 46 candidates through visual inspection and then used the Gemini Multi-Object Spectrograph (GMOS) at the Gemini South telescope to acquire spectroscopic follow-up of 21 of these candidates. Through analysis of this spectroscopic follow-up data, we confirmed nine new lensing systems and rejected 2 candidates, but the analysis was inconclusive on 10 candidates. For each of the confirmed systems, we report measured spectroscopic properties, estimated \einsteinradiussub, and estimated enclosed masses. The sources that we targeted have an i-band surface brightness range of iSB ~ 22 - 24 mag arcsec^2 and a spectroscopic redshift range of zspec ~0.8 - 2.6. The lens galaxies have a photometric redshift range of zlens ~ 0.3 - 0.7. The lensing systems range in image-lens separation 2 - 9 arcsec and in enclosed mass 10^12 - 10^13 Msol.

astro-ph.CO↗

arXiv2019

Response to NITRD, NCO, NSF Request for Information on "Update to the 2016 National Artificial Intelligence Research and Development Strategic Plan"

We present a response to the 2018 Request for Information (RFI) from the NITRD, NCO, NSF regarding the "Update to the 2016 National Artificial Intelligence Research and Development Strategic Plan." Through this document, we provide a response to the question of whether and how the National Artificial Intelligence Research and Development Strategic Plan (NAIRDSP) should be updated from the perspective of Fermilab, America's premier national laboratory for High Energy Physics (HEP). We believe the NAIRDSP should be extended in light of the rapid pace of development and innovation in the field of Artificial Intelligence (AI) since 2016, and present our recommendations below. AI has profoundly impacted many areas of human life, promising to dramatically reshape society --- e.g., economy, education, science --- in the coming years. We are still early in this process. It is critical to invest now in this technology to ensure it is safe and deployed ethically. Science and society both have a strong need for accuracy, efficiency, transparency, and accountability in algorithms, making investments in scientific AI particularly valuable. Thus far the US has been a leader in AI technologies, and we believe as a national Laboratory it is crucial to help maintain and extend this leadership. Moreover, investments in AI will be important for maintaining US leadership in the physical sciences.

astro-ph.IM↗

arXiv2019

First Cosmology Results Using Type Ia Supernovae From the Dark Energy Survey: Photometric Pipeline and Light Curve Data Release

We present griz light curves of 251 Type Ia Supernovae (SNe Ia) from the first 3 years of the Dark Energy Survey Supernova Program's (DES-SN) spectroscopically classified sample. The photometric pipeline described in this paper produces the calibrated fluxes and associated uncertainties used in the cosmological parameter analysis (Brout et al. 2018-SYS, DES Collaboration et al. 2018) by employing a scene modeling approach that simultaneously forward models a variable transient flux and temporally constant host galaxy. We inject artificial point sources onto DECam images to test the accuracy of our photometric method. Upon comparison of input and measured artificial supernova fluxes, we find flux biases peak at 3 mmag. We require corrections to our photometric uncertainties as a function of host galaxy surface brightness at the transient location, similar to that seen by the DES Difference Imaging Pipeline used to discover transients. The public release of the light curves can be found at https://des.ncsa.illinois.edu/releases/sn.

astro-ph.IM↗

arXiv2019

First Cosmology Results using Type Ia Supernova from the Dark Energy Survey: Simulations to Correct Supernova Distance Biases

We describe catalog-level simulations of Type Ia supernova (SN~Ia) light curves in the Dark Energy Survey Supernova Program (DES-SN), and in low-redshift samples from the Center for Astrophysics (CfA) and the Carnegie Supernova Project (CSP). These simulations are used to model biases from selection effects and light curve analysis, and to determine bias corrections for SN~Ia distance moduli that are used to measure cosmological parameters. To generate realistic light curves, the simulation uses a detailed SN~Ia model, incorporates information from observations (PSF, sky noise, zero point), and uses summary information (e.g., detection efficiency vs. signal to noise ratio) based on 10,000 fake SN light curves whose fluxes were overlaid on images and processed with our analysis pipelines. The quality of the simulation is illustrated by predicting distributions observed in the data. Averaging within redshift bins, we find distance modulus biases up to 0.05~mag over the redshift ranges of the low-z and DES-SN samples. For individual events, particularly those with extreme red or blue color, distance biases can reach 0.4~mag. Therefore, accurately determining bias corrections is critical for precision measurements of cosmological parameters. Files used to make these corrections are available at https://des.ncsa.illinois.edu/releases/sn.

astro-ph.CO↗