arXiv subjects

D. Cornu

Publications and source records attributed to D. Cornu.

9 recordsLinked to original sources

arXiv2025

YOLO-CIANNA: Galaxy detection with deep learning in radio data: II. Winning the SKA SDC2 using a generalized 3D-YOLO network

As the scientific exploitation of the Square Kilometre Array (SKA) approaches, there is a need for new advanced data analysis and visualization tools capable of processing large high-dimensional datasets. In this study, we aim to generalize the YOLO-CIANNA deep learning source detection and characterization method for 3D hyperspectral HI emission cubes. We present the adaptations we made to the regression-based detection formalism and the construction of an end-to-end 3D convolutional neural network (CNN) backbone. We then describe a processing pipeline for applying the method to simulated 3D HI cubes from the SKA Observatory Science Data Challenge 2 (SDC2) dataset. The YOLO-CIANNA method was originally developed and used by the MINERVA team that won the official SDC2 competition. Despite the public release of the full SDC2 dataset, no published result has yet surpassed MINERVA's top score. In this paper, we present an updated version of our method that improves our challenge score by 9.5%. The resulting catalog exhibits a high detection purity of 92.3%, best-in-class characterization accuracy, and contains 45% more confirmed sources than concurrent classical detection tools. The method is also computationally efficient, processing the full ~1TB SDC2 data cube in 30 min on a single GPU. These state-of-the-art results highlight the effectiveness of 3D CNN-based detectors for processing large hyperspectral data cubes and represent a promising step toward applying YOLO-CIANNA to observational data from SKA and its precursors.

astro-ph.GA↗

arXiv2025

Square Kilometre Array Science Data Challenge 3a: foreground removal for an EoR experiment

We present and analyse the results of the Science data challenge 3a (SDC3a, https://sdc3.skao.int/challenges/foregrounds), an EoR foreground-removal community-wide exercise organised by the Square Kilometre Array Observatory (SKAO). The challenge ran for 8 months, from March to October 2023. Participants were provided with realistic simulations of SKA-Low data between 106 MHz and 196 MHz, including foreground contamination from extragalactic as well as Galactic emission, instrumental and systematic effects. They were asked to deliver cylindrical power spectra of the EoR signal, cleaned from all corruptions, and the corresponding confidence levels. Here we describe the approaches taken by the 17 teams that completed the challenge, and we assess their performance using different metrics. The challenge results provide a positive outlook on the capabilities of current foreground-mitigation approaches to recover the faint EoR signal from SKA-Low observations. The median error committed in the EoR power spectrum recovery is below the true signal for seven teams, although in some cases there are some significant outliers. The smallest residual overall is $4.2_{-4.2}^{+20} \times 10^{-4}\,\rm{K}^2h^{-3}$cMpc$^{3}$ across all considered scales and frequencies. The estimation of confidence levels provided by the teams is overall less accurate, with the true error being typically under-estimated, sometimes very significantly. The most accurate error bars account for $60 \pm 20$\% of the true errors committed. The challenge results provide a means for all teams to understand and improve their performance. This challenge indicates that the comparison between independent pipelines could be a powerful tool to assess residual biases and improve error estimation.

astro-ph.IM↗

arXiv2024

The Galaxy Activity, Torus, and Outflow Survey (GATOS). Black hole mass estimation using machine learning

The detailed feeding and feedback mechanisms of Active Galactic Nuclei (AGN) are not yet well known. For low-luminosity and obscured AGN, as well as late-type galaxies, determining the central black hole (BH) masses is challenging. Our goal with the GATOS sample is to study circum-nuclear regions and better estimate BH masses with more precision than scaling relations offer. Using ALMA's high spatial resolution, we resolve CO(3-2) emissions within ~100 pc around the supermassive black hole (SMBH) in seven GATOS galaxies to estimate their BH masses when sufficient gas is present. We study seven bright ($L_{AGN}(14-150\mathrm{keV}) \geq 10^{42}\mathrm{erg/s}$), nearby (<28 Mpc) galaxies from the GATOS core sample. For comparison, we searched the literature for previous BH mass estimates and made additional calculations using the \mbh~ - $\sigma$ relation and the fundamental plane of BH activity. We developed a supervised machine learning method to estimate BH masses from position-velocity diagrams or first-moment maps using ALMA CO(3-2) observations. Numerical simulations with a wide range of parameters created the training, validation, and test sets. Seven galaxies provided enough gas for BH mass estimations: NGC4388, NGC5506, NGC5643, NGC6300, NGC7314, NGC7465, and NGC~7582. Our BH masses, ranging from 6.39 to 7.18 log$(M_{BH}/M_\odot)$, align with previous estimates. Additionally, our machine learning method provides robust error estimations with confidence intervals and offers greater potential than scaling relations. This work is a first step toward an automated \mbh estimation method using machine learning.

astro-ph.GA↗

arXiv2024

YOLO-CIANNA: Galaxy detection with deep learning in radio data. I. A new YOLO-inspired source detection method applied to the SKAO SDC1

The upcoming Square Kilometer Array (SKA) will set a new standard regarding data volume generated by an astronomical instrument, which is likely to challenge widely adopted data-analysis tools that scale inadequately with the data size. The aim of this study is to develop a new source detection and characterization method for massive radio astronomical datasets based on modern deep-learning object detection techniques. For this, we seek to identify the specific strengths and weaknesses of this type of approach when applied to astronomical data. We introduce YOLO-CIANNA, a highly customized deep-learning object detector designed specifically for astronomical datasets. In this paper, we present the method and describe all the elements introduced to address the specific challenges of radio astronomical images. We then demonstrate the capabilities of this method by applying it to simulated 2D continuum images from the SKA observatory Science Data Challenge 1 (SDC1) dataset. Using the SDC1 metric, we improve the challenge-winning score by +139% and the score of the only other post-challenge participation by +61%. Our catalog has a detection purity of 94% while detecting 40 to 60% more sources than previous top-score results, and exhibits strong characterization accuracy. The trained model can also be forced to reach 99% purity in post-process and still detect 10 to 30% more sources than the other top-score methods. It is also computationally efficient, with a peak prediction speed of 500 images of 512x512 pixels per second on a single GPU. YOLO-CIANNA achieves state-of-the-art detection and characterization results on the simulated SDC1 dataset and is expected to transfer well to observational data from SKA precursors.

astro-ph.IM↗

arXiv2023

SKA Science Data Challenge 2: analysis and results

The Square Kilometre Array Observatory (SKAO) will explore the radio sky to new depths in order to conduct transformational science. SKAO data products made available to astronomers will be correspondingly large and complex, requiring the application of advanced analysis techniques to extract key science findings. To this end, SKAO is conducting a series of Science Data Challenges, each designed to familiarise the scientific community with SKAO data and to drive the development of new analysis techniques. We present the results from Science Data Challenge 2 (SDC2), which invited participants to find and characterise 233245 neutral hydrogen (Hi) sources in a simulated data product representing a 2000~h SKA MID spectral line observation from redshifts 0.25 to 0.5. Through the generous support of eight international supercomputing facilities, participants were able to undertake the Challenge using dedicated computational resources. Alongside the main challenge, `reproducibility awards' were made in recognition of those pipelines which demonstrated Open Science best practice. The Challenge saw over 100 participants develop a range of new and existing techniques, with results that highlight the strengths of multidisciplinary and collaborative effort. The winning strategy -- which combined predictions from two independent machine learning techniques to yield a 20 percent improvement in overall performance -- underscores one of the main Challenge outcomes: that of method complementarity. It is likely that the combination of methods in a so-called ensemble approach will be key to exploiting very large astronomical datasets.

astro-ph.IM↗

arXiv2022

3D extinction mapping of the Milky Way using Convolutional Neural Networks: Presentation of the method and demonstration in the Carina Arm region

Context. Several methods have been proposed to build 3D extinction maps of the Milky Way (MW), most often based on Bayesian approaches. Although some studies employed machine learning (ML) methods in part of their procedure, or to specific targets, no 3D extinction map of a large volume of the MW solely based on a Neural Network method has been reported so far. Aims. We aim to apply deep learning as a solution to build 3D extinction maps of the MW. Methods. We built a convolutional neural network (CNN) using the CIANNA framework, and trained it with synthetic 2MASS data. We used the Besan\c{c}on Galaxy model to generate mock star catalogs, and 1D Gaussian random fields to simulate the extinction profiles. From these data we computed color-magnitude diagrams (CMDs) to train the network, using the corresponding extinction profiles as targets. A forward pass with observed 2MASS CMDs provided extinction profile estimates for a grid of lines of sight. Results. We trained our network with data simulating lines of sight in the area of the Carina spiral arm tangent and obtained a 3D extinction map for a large sector in this region ($l = 257 - 303$ deg, $|b| \le 5$ deg), with distance and angular resolutions of $100$ pc and $30$ arcmin, respectively, and reaching up to $\sim 10$ kpc. Although each sightline is computed independently in the forward phase, the so-called fingers-of-God artifacts are weaker than in many other 3D extinction maps. We found that our CNN was efficient in taking advantage of redundancy across lines of sight, enabling us to train it with only 9 sightlines simultaneously to build the whole map. Conclusions. We found deep learning to be a reliable approach to produce 3D extinction maps from large surveys. With this methodology, we expect to easily combine heterogeneous surveys without cross-matching, and therefore to exploit several surveys in a complementary fashion.

astro-ph.GA↗

arXiv2021

Deciphering the evolution of the Milky Way discs: Gaia APOGEE Kepler giant stars and the Besan\c{c}on Galaxy Model

We investigate the properties of the double sequences of the Milky Way discs visible in the [$\alpha$/Fe] vs [Fe/H] diagram. In the framework of Galactic formation and evolution, we discuss the complex relationships between age, metallicity, [$\alpha$/Fe], and the velocity components. We study stars with measured chemical, seismic and astrometric properties from the APOGEE survey, the Kepler and Gaia satellites, respectively. We separate the [$\alpha$/Fe]-[Fe/H] diagram into 3 stellar populations: the thin disc, the high-$\alpha$ metal-poor thick disc and the high-$\alpha$ metal-rich thick disc and characterise each of these in the age-chemo-kinematics parameter space. We compare results obtained from different APOGEE data releases and using two recent age determinations. We use the Besan\c{c}on Galaxy model (BGM) to highlight selection biases and mechanisms not included in the model. The thin disc exhibits a flat age-metallicity relation while [$\alpha$/Fe] increases with stellar age. We confirm no correlation between radial and vertical velocities with [Fe/H], [$\alpha$/Fe] and age for each stellar population. Considering both samples, V$_\varphi$ decreases with age for the thin disc, while it increases with age for the h$\alpha$mp thick disc. Although the age distribution of the h$\alpha$mr thick disc is very close to that of the h$\alpha$mp thick disc between 7 and 14 Gyr, its kinematics seems to follow that of the thin disc. This feature, not predicted by the hypotheses included in the BGM, suggests a different origin and history for this population. Finally, we show that there is a maximum dispersion of the vertical velocity, $\sigma_Z$, with age for the h$\alpha$mp thick disc around 8 Gyr. The comparisons with the BGM simulations suggest a more complex chemo-dynamical scheme to explain this feature, most likely including mergers and radial migration effects

astro-ph.GA↗

arXiv2019

SCOPE: SCUBA-2 Continuum Observations of Pre-protostellar Evolution - Survey Description and Compact Source Catalogue

We present the first release of the data and compact-source catalogue for the JCMT Large Program SCUBA-2 Continuum Observations of Pre-protostellar Evolution (SCOPE). SCOPE consists of 850-um continuum observations of 1235 Planck Galactic Cold Clumps (PGCCs) made with the Submillimetre Common-User Bolometer Array 2 on the James Clerk Maxwell Telescope. These data are at an angular resolution of 14.4 arcsec, significantly improving upon the 353-GHz resolution of Planck at 5 arcmin, and allowing for a catalogue of 3528 compact sources in 558 PGCCs. We find that the detected PGCCs have significant sub-structure, with 61 per cent of detected PGCCs having 3 or more compact sources, with filamentary structure also prevalent within the sample. A detection rate of 45 per cent is found across the survey, which is 95 per cent complete to Planck column densities of $N_{H_{2}}$ $>$ 5 $\times$ 10$^{21}$ cm$^{-2}$. By positionally associating the SCOPE compact sources with YSOs, the star formation efficiency, as measured by the ratio of luminosity to mass, in nearby clouds is found to be similar to that in the more distant Galactic Plane, with the column density distributions also indistinguishable from each other.

astro-ph.GA↗

arXiv2010

Self-oscillations in field emission nanowire mechanical resonators: a nanometric DC-AC conversion

We report the observation of self-oscillations in a bottom-up nanoelectromechanical system (NEMS) during field emission driven by a constant applied voltage. An electromechanical model is explored that explains the phenomenon and that can be directly used to develop integrated devices. In this first study we have already achieved ~50% DC/AC (direct to alternative current) conversion. Electrical self-oscillations in NEMS open up a new path for the development of high speed, autonomous nanoresonators, and signal generators and show that field emission (FE) is a powerful tool for building new nano-components.

cond-mat.mes-hall↗