arXiv subjects

E. Russeil

Publications and source records attributed to E. Russeil.

9 recordsLinked to original sources

arXiv2026

NOMAI : A real-time photometric classifier for superluminous supernovae identification. A science module for the Fink broker

Superluminous supernovae (SLSNe) are one of the most luminous stellar explosions known, yet they remain poorly understood. Because they are intrinsically rare, efficiently identifying them in the large alert streams produced by modern time-domain surveys is essential for enabling spectroscopic follow-up. We present NOMAI, a machine learning classifier designed to identify SLSN candidates directly from photometric alerts in the ZTF stream, using light curves accumulated over at least 30 days. It does not require any spectroscopic redshift and is running in real time within the Fink broker. ZTF light curves are transformed into a set of physically motivated features derived primarily from model-fitting procedures using SALT2 and Rainbow, a blackbody-based multi-band fitting framework. These features are used to train an XGBoost classifier on a curated dataset of labeled ZTF sources constructed using literature samples of SLSNe, along with TNS and internal ZTF labeled sources. The final training dataset contains 5280 unique sources, including 225 spectroscopically classified SLSNe. On the training sample, the classifier reaches 66% completeness and 58% purity. Deployed within the Fink broker, NOMAI has been running continuously since 18/12/2025 on the ZTF alert stream and publicly reports SLSN candidates every night by automatically posting them to dedicated communication channels. Based on this, we also report the first two-month as an evaluation period, where the classifier successfully recovered 22 of the 24 active SLSNe reported on the Transient Name Server. The achieved performances demonstrate that the classifier provides a valuable tool for experts to efficiently scan the alert stream and identify promising candidates. In the near future, NOMAI is intended to be adapted to operate on the Legacy Survey of Space and Time conducted by the Vera C. Rubin Observatory.

astro-ph.IM↗

arXiv2025

Lost and Found - A gallery of overlooked optical nuclear transients from the ZTF archive

Tidal disruption events (TDEs) correspond to the destruction of a star by the tidal forces around a black hole, leading to outbursts which can last from months to years. These transients are rare, and increasing the current sample is paramount to understand them. As part of the Fink alert broker, we have developed an early detection system for TDEs for the Zwicky Transient Facility (ZTF) data. In this paper, we report on the optical transients we found either during the development of this tool, or when applying the classifier to the existing archive. We use this sample to anticipate what improvements to the TDE detection systems will need to be implemented for future surveys. For all the transients, we present optical and infrared archival photometry from ZTF, WISE, and Catalina, and assess the previous nuclear activity of the host. We fit the ZTF lightcurves with both a phenomenological and a physically-motivated model. We report on a total of 19 optical nuclear transients, out of which nine are in passive galaxies, eight in active galaxies, and two for which the activity of the host is uncertain. Two transients are newly discovered repeated TDE candidates, and we compare them to the current sample of repeated optical nuclear transients. One transient is exceptionally long-lived (over 5 years), in an until-now passive galaxy. Three of the TDE-like flares in active galaxies have absolute g-band magnitudes brighter than -24, making them new Extreme Nuclear Transient (ENT) candidates. One seemingly repeated object was revealed to be two independent supernovae in the same galaxy. This sample shows both the potential of our detection system for future discovery, and the relevance of archival searches to reveal overlooked transients. It also raises several points of concern and avenues of improvement for current and future classifiers.

astro-ph.HE↗

arXiv2025

Eruptive mass loss less than a year before the explosion of superluminous supernovae. II. A systematic search for pre-explosion eruptions with VLT/X-shooter

We present X-shooter observations of a sample of 21 hydrogen-poor superluminous supernovae (SLSNe-I), spanning a redshift range of z=0.13-0.95, aimed at searching for shells of circumstellar material (CSM). Specifically, we focus on identifying broad Mg II absorption features that are blueshifted by several thousand kilometers per second and have previously been interpreted as arising from resonance line scattering of the SLSN continuum by rapidly expanding CSM ejected shortly before explosion. Utilizing high-quality spectra, we model the region around 2800A to characterize the Mg II line profiles, enabling us to either confirm their presence or place constraints on undetected CSM shells. We identify five objects in our sample that show broad Mg II absorption features consistent with the presence of CSM. While SN2018ibb, SN2020xga and SN2022xgc have been previously reported, we identify previously undiscovered CSM shells in DES15S2nr and DES16C3ggu. These shells were likely expelled approximately two and three months, respectively, before the explosion of their associated SNe, timescales consistent with late-stage mass-loss episodes. We do not find any correlations between the shell properties and the SN properties, except for a marginal correlation between the light curve decline time scale and the shell velocities. We further demonstrate that CSM configurations similar to the majority of the detected shells would have been observable in spectra with signal-to-noise >8 per resolution element, and that the lines from a shell are in general detectable except in the cases where the shell is either very geometrically and/or optically thin. Therefore, we conclude that the detection of CSM shells is not a selection effect, but may instead point to the existence of a subclass of SLSNe-I undergoing late-stage shell ejections shortly before explosion.

astro-ph.HE↗

arXiv2024

Coniferest: a complete active anomaly detection framework

We present coniferest, an open source generic purpose active anomaly detection framework written in Python. The package design and implemented algorithms are described. Currently, static outlier detection analysis is supported via the Isolation forest algorithm. Moreover, Active Anomaly Discovery (AAD) and Pineforest algorithms are available to tackle active anomaly detection problems. The algorithms and package performance are evaluated on a series of synthetic datasets. We also describe a few success cases which resulted from applying the package to real astronomical data in active anomaly detection tasks within the SNAD project.

astro-ph.IM↗

arXiv2024

Real-bogus scores for active anomaly detection

In the task of anomaly detection in modern time-domain photometric surveys, the primary goal is to identify astrophysically interesting, rare, and unusual objects among a large volume of data. Unfortunately, artifacts -- such as plane or satellite tracks, bad columns on CCDs, and ghosts -- often constitute significant contaminants in results from anomaly detection analysis. In such contexts, the Active Anomaly Discovery (AAD) algorithm allows tailoring the output of anomaly detection pipelines according to what the expert judges to be scientifically interesting. We demonstrate how the introduction real-bogus scores, obtained from a machine learning classifier, improves the results from AAD. Using labeled data from the SNAD ZTF knowledge database, we train four real-bogus classifiers: XGBoost, CatBoost, Random Forest, and Extremely Randomized Trees. All the models perform real-bogus classification with similar effectiveness, achieving ROC-AUC scores ranging from 0.93 to 0.95. Consequently, we select the Random Forest model as the main model due to its simplicity and interpretability. The Random Forest classifier is applied to 67 million light curves from ZTF DR17. The output real-bogus score is used as an additional feature for two anomaly detection algorithms: static Isolation Forest and AAD. While results from Isolation Forest remained unchanged, the number of artifacts detected by the active approach decreases significantly with the inclusion of the real-bogus score, from 27 to 3 out of 100. We conclude that incorporating the real-bogus classifier result as an additional feature in the active anomaly detection pipeline significantly reduces the number of artifacts in the outputs, thereby increasing the incidence of astrophysically interesting objects presented to human experts.

astro-ph.IM↗

arXiv2024

Transient Classifiers for Fink: Benchmarks for LSST

The upcoming Legacy Survey of Space and Time (LSST) is expected to detect a few million transients per night, which will generate a live alert stream during the entire ten years of the survey. This stream will be distributed via community brokers whose task is to select subsets of the stream and direct them to scientific communities. Given the volume and complexity of the anticipated data, machine learning algorithms will be paramount for this task. We present the infrastructure tests and classification methods developed within the Fink broker in preparation for LSST. This work aims to provide detailed information regarding the underlying assumptions and methods behind each classifier and enable users to make informed follow-up decisions from Fink photometric classifications. Using simulated data from ELAsTiCC, we showcase the performance of binary and multi-class ML classifiers available in Fink. These include tree-based classifiers coupled with tailored feature extraction strategies as well as deep learning algorithms. Moreover, we introduce CATS, a deep learning architecture specifically designed for this task. Our results show that Fink classifiers are able to handle the extra complexity that is expected from LSST data. CATS achieved $\geq 93\%$ precision for all classes except `long' (for which it achieved $\sim 83\%$), while our best performing binary classifier achieves $\geq 98\%$ precision and $\geq 99\%$ completeness when classifying the periodic class. ELAsTiCC was an important milestone in preparing the Fink infrastructure to deal with LSST-like data. Our results demonstrate that Fink classifiers are well prepared for the arrival of the new stream, but this work also highlights that transitioning from the current infrastructures to Rubin will require significant adaptation of the currently available tools. This work was the first step in the right direction.

astro-ph.IM↗

arXiv2024

SNAD catalogue of M-dwarf flares from the Zwicky Transient Facility

Most of the stars in the Universe are M spectral class dwarfs, which are known to be the source of bright and frequent stellar flares. In this paper, we propose new approaches to discover M-dwarf flares in ground-based photometric surveys. We employ two approaches: a modification of a traditional method of parametric fit search and a machine learning algorithm based on active anomaly detection. The algorithms are applied to Zwicky Transient Facility (ZTF) data release 8, which includes the data from the ZTF high-cadence survey, allowing us to reveal flares lasting from minutes to hours. We analyze over 35 million ZTF light curves and visually scrutinize 1168 candidates suggested by the algorithms to filter out artifacts, occultations of a star by an asteroid, and other types of known variable objects. The result of this analysis is the largest catalogue of ZTF flaring stars to date, representing 134 flares with amplitudes ranging from -0.2 to -4.6 magnitudes, including repeated flares. Using Pan-STARRS DR2 colors, we assign a spectral subclass to each object in the sample. For 13 flares with well-sampled light curves and available geometric distances from Gaia DR3, we estimate the bolometric energy. This research shows that the proposed methods combined with the ZTF's cadence strategy are suitable for identifying M-dwarf flares and other fast transients, allowing for the extraction of significant astrophysical information from their light curves.

astro-ph.SR↗

arXiv2023

Rainbow: a colorful approach on multi-passband light curve estimation

We present Rainbow, a physically motivated framework which enables simultaneous multi-band light curve fitting. It allows the user to construct a 2-dimensional continuous surface across wavelength and time, even in situations where the number of observations in each filter is significantly limited. Assuming the electromagnetic radiation emission from the transient can be approximated by a black-body, we combined an expected temperature evolution and a parametric function describing its bolometric light curve. These three ingredients allow the information available in one passband to guide the reconstruction in the others, thus enabling a proper use of multi-survey data. We demonstrate the effectiveness of our method by applying it to simulated data from the Photometric LSST Astronomical Time-series Classification Challenge (PLAsTiCC) as well as real data from the Young Supernova Experiment (YSE DR1). We evaluate the quality of the estimated light curves according to three different tests: goodness of fit, time of peak prediction and ability to transfer information to machine learning (ML) based classifiers. Results confirm that Rainbow leads to equivalent (SNII) or up to 75% better (SN Ibc) goodness of fit when compared to the Monochromatic approach. Similarly, accuracy when using Rainbow best-fit values as a parameter space in multi-class ML classification improves for all classes in our sample. An efficient implementation of Rainbow has been publicly released as part of the light curve package at https://github.com/light-curve/light-curve-python. Our approach enables straight forward light curve estimation for objects with observations in multiple filters and from multiple experiments. It is particularly well suited for situations where light curve sampling is sparse.

astro-ph.IM↗

arXiv2022

SNAD Transient Miner: Finding Missed Transient Events in ZTF DR4 using k-D trees

We report the automatic detection of 11 transients (7 possible supernovae and 4 active galactic nuclei candidates) within the Zwicky Transient Facility fourth data release (ZTF DR4), all of them observed in 2018 and absent from public catalogs. Among these, three were not part of the ZTF alert stream. Our transient mining strategy employs 41 physically motivated features extracted from both real light curves and four simulated light curve models (SN Ia, SN II, TDE, SLSN-I). These features are input to a k-D tree algorithm, from which we calculate the 15 nearest neighbors. After pre-processing and selection cuts, our dataset contained approximately a million objects among which we visually inspected the 105 closest neighbors from seven of our brightest, most well-sampled simulations, comprising 89 unique ZTF DR4 sources. Our result illustrates the potential of coherently incorporating domain knowledge and automatic learning algorithms, which is one of the guiding principles directing the SNAD team. It also demonstrates that the ZTF DR is a suitable testing ground for data mining algorithms aiming to prepare for the next generation of astronomical data.

astro-ph.IM↗