arXiv subjects

Konstantin Malanchev

Publications and source records attributed to Konstantin Malanchev.

35 records · Page 2Linked to original sources

arXiv2024

Multi-View Symbolic Regression

Symbolic regression (SR) searches for analytical expressions representing the relationship between a set of explanatory and response variables. Current SR methods assume a single dataset extracted from a single experiment. Nevertheless, frequently, the researcher is confronted with multiple sets of results obtained from experiments conducted with different setups. Traditional SR methods may fail to find the underlying expression since the parameters of each experiment can be different. In this work we present Multi-View Symbolic Regression (MvSR), which takes into account multiple datasets simultaneously, mimicking experimental environments, and outputs a general parametric solution. This approach fits the evaluated expression to each independent dataset and returns a parametric family of functions f(x; theta) simultaneously capable of accurately fitting all datasets. We demonstrate the effectiveness of MvSR using data generated from known expressions, as well as real-world data from astronomy, chemistry and economy, for which an a priori analytical expression is not available. Results show that MvSR obtains the correct expression more frequently and is robust to hyperparameters change. In real-world data, it is able to grasp the group behavior, recovering known expressions from the literature as well as promising alternatives, thus enabling the use of SR to a large range of experimental scenarios.

cs.LG↗

arXiv2024

Exploring the Universe with SNAD: Anomaly Detection in Astronomy

SNAD is an international project with a primary focus on detecting astronomical anomalies within large-scale surveys, using active learning and other machine learning algorithms. The work carried out by SNAD not only contributes to the discovery and classification of various astronomical phenomena but also enhances our understanding and implementation of machine learning techniques within the field of astrophysics. This paper provides a review of the SNAD project and summarizes the advancements and achievements made by the team over several years.

astro-ph.IM↗

arXiv2024

Superphot+: Realtime Fitting and Classification of Supernova Light Curves

Photometric classifications of supernova (SN) light curves have become necessary to utilize the full potential of large samples of observations obtained from wide-field photometric surveys, such as the Zwicky Transient Facility (ZTF) and the Vera C. Rubin Observatory. Here, we present a photometric classifier for SN light curves that does not rely on redshift information and still maintains comparable accuracy to redshift-dependent classifiers. Our new package, Superphot+, uses a parametric model to extract meaningful features from multiband SN light curves. We train a gradient-boosted machine with fit parameters from 6,061 ZTF SNe that pass data quality cuts and are spectroscopically classified as one of five classes: SN Ia, SN II, SN Ib/c, SN IIn, and SLSN-I. Without redshift information, our classifier yields a class-averaged F1-score of 0.61 +/- 0.02 and a total accuracy of 0.83 +/- 0.01. Including redshift information improves these metrics to 0.71 +/- 0.02 and 0.88 +/- 0.01, respectively. We assign new class probabilities to 3,558 ZTF transients that show SN-like characteristics (based on the ALeRCE Broker light curve and stamp classifiers), but lack spectroscopic classifications. Finally, we compare our predicted SN labels with those generated by the ALeRCE light curve classifier, finding that the two classifiers agree on photometric labels for 82 +/- 2% of light curves with spectroscopic labels and 72% of light curves without spectroscopic labels. Superphot+ is currently classifying ZTF SNe in real time via the ANTARES Broker, and is designed for simple adaptation to six-band Rubin light curves in the future.

astro-ph.HE↗

arXiv2023

Automatic detection of plateau phases in light curves of variable stars

Modern astronomical surveys produce millions of light curves of variable sources. These massive data sets challenge the community to create automatic light-curve processing methods for detection, classification, and characterisation of variable stars. In this paper, we present a novel method for extracting the variable components of a light curve based on Otsu's thresholding method. To validate the effectiveness of this method, we apply it to the light curves of detached eclipsing binaries and dwarf novae, sourced from OGLE catalogues.

astro-ph.IM↗

arXiv2023

Understanding of the properties of neural network approaches for transient light curve approximations

Modern-day time-domain photometric surveys collect a lot of observations of various astronomical objects and the coming era of large-scale surveys will provide even more information on their properties. Spectroscopic follow-ups are especially crucial for transients such as supernovae and most of these objects have not been subject to such studies. }{Flux time series are actively used as an affordable alternative for photometric classification and characterization, for instance, peak identifications and luminosity decline estimations. However, the collected time series are multidimensional and irregularly sampled, while also containing outliers and without any well-defined systematic uncertainties. This paper presents a search for the best-performing methods to approximate the observed light curves over time and wavelength for the purpose of generating time series with regular time steps in each passband.}{We examined several light curve approximation methods based on neural networks such as multilayer perceptrons, Bayesian neural networks, and normalizing flows to approximate observations of a single light curve. Test datasets include simulated PLAsTiCC and real Zwicky Transient Facility Bright Transient Survey light curves of transients.}{The tests demonstrate that even just a few observations are enough to fit the networks and improve the quality of approximation, compared to state-of-the-art models. The methods described in this work have a low computational complexity and are significantly faster than Gaussian processes. Additionally, we analyzed the performance of the approximation techniques from the perspective of further peak identification and transients classification. The study results have been released in an open and user-friendly Fulu Python library available on GitHub for the scientific community.

astro-ph.IM↗

arXiv2023

Applications of Deep Learning to physics workflows

Modern large-scale physics experiments create datasets with sizes and streaming rates that can exceed those from industry leaders such as Google Cloud and Netflix. Fully processing these datasets requires both sufficient compute power and efficient workflows. Recent advances in Machine Learning (ML) and Artificial Intelligence (AI) can either improve or replace existing domain-specific algorithms to increase workflow efficiency. Not only can these algorithms improve the physics performance of current algorithms, but they can often be executed more quickly, especially when run on coprocessors such as GPUs or FPGAs. In the winter of 2023, MIT hosted the Accelerating Physics with ML at MIT workshop, which brought together researchers from gravitational-wave physics, multi-messenger astrophysics, and particle physics to discuss and share current efforts to integrate ML tools into their workflows. The following white paper highlights examples of algorithms and computing frameworks discussed during this workshop and summarizes the expected computing needs for the immediate future of the involved fields.

hep-ex↗

arXiv2023

All-Sky Faint DA White Dwarf Spectrophotometric Standards for Astrophysical Observatories: The Complete Sample

Hot DA white dwarfs have fully radiative pure hydrogen atmospheres that are the least complicated to model. Pulsationally stable, they are fully characterized by their effective temperature Teff, and surface gravity log g, which can be deduced from their optical spectra and used in model atmospheres to predict their spectral energy distribution (SED). Based on this, three bright DAWDs have defined the spectrophotometric flux scale of the CALSPEC system of HST. In this paper we add 32 new fainter (16.5 < V < 19.5) DAWDs spread over the whole sky and within the dynamic range of large telescopes. Using ground based spectra and panchromatic photometry with HST/WFC3, a new hierarchical analysis process demonstrates consistency between model and observed fluxes above the terrestrial atmosphere to < 0.004 mag rms from 2700 Å to 7750 Å and to 0.008 mag rms at 1.6μm for the total set of 35 DAWDs. These DAWDs are thus established as spectrophotometric standards with unprecedented accuracy from the near ultraviolet to the near-infrared, suitable for both ground and space based observatories. They are embedded in existing surveys like SDSS, PanSTARRS and GAIA, and will be naturally included in the LSST survey by Rubin Observatory. With additional data and analysis to extend the validity of their SEDs further into the IR, these spectrophotometric standard stars could be used for JWST, as well as for the Roman and Euclid observatories.

astro-ph.SR↗

arXiv2023

The SNAD Viewer: Everything You Want to Know about Your Favorite ZTF Object

We describe the SNAD Viewer, a web portal for astronomers which presents a centralized view of individual objects from the Zwicky Transient Facility's (ZTF) data releases, including data gathered from multiple publicly available astronomical archives and data sources. Initially built to enable efficient expert feedback in the context of adaptive machine learning applications, it has evolved into a full-fledged community asset that centralizes public information and provides a multi-dimensional view of ZTF sources. For users, we provide detailed descriptions of the data sources and choices underlying the information displayed in the portal. For developers, we describe our architectural choices and their consequences such that our experience can help others engaged in similar endeavors or in adapting our publicly released code to their requirements. The infrastructure we describe here is scalable and flexible and can be personalized and used by other surveys and for other science goals. The Viewer has been instrumental in highlighting the crucial roles domain experts retain in the era of big data in astronomy. Given the arrival of the upcoming generation of large-scale surveys, we believe similar systems will be paramount in enabling an optimal exploitation of the scientific potential enclosed in current terabyte and future petabyte-scale data sets. The Viewer is publicly available online at https://ztf.snad.space

astro-ph.IM↗

arXiv2023

Performant feature extraction for photometric time series

Astronomy is entering the era of large surveys of the variable sky such as Zwicky Transient Facility (ZTF) and the forthcoming Legacy Survey of Space and Time (LSST) which are intended to produce up to a million alerts per night. Such an amount of photometric data requires efficient light-curve pre-processing algorithms for the purposes of subsequent data quality cuts, classification, and characterization analysis. In this work, we present the new library "light-curve" for Python and Rust, which is intended for feature extraction from light curves of variable astronomical sources. The library is suitable for machine learning classification problems: it provides a fast implementation of feature extractors, which outperforms other public available codes, and consists of dozens features describing shape, magnitude distribution, and periodic properties of light curves. It includes not only features which had been shown to provide a high performance in classification tasks, but also new features we developed to improve classification quality of selected types of objects. The "light-curve" library is currently used by the ANTARES, AMPEL, and Fink broker systems for analyzing the ZTF alert stream, and has been selected for use with the LSST.

astro-ph.IM↗

arXiv2022

Rubin Observatory LSST Transients and Variable Stars Roadmap

The Vera C. Rubin Legacy Survey of Space and Time holds the potential to revolutionize time domain astrophysics, reaching completely unexplored areas of the Universe and mapping variability time scales from minutes to a decade. To prepare to maximize the potential of the Rubin LSST data for the exploration of the transient and variable Universe, one of the four pillars of Rubin LSST science, the Transient and Variable Stars Science Collaboration, one of the eight Rubin LSST Science Collaborations, has identified research areas of interest and requirements, and paths to enable them. While our roadmap is ever-evolving, this document represents a snapshot of our plans and preparatory work in the final years and months leading up to the survey's first light.

astro-ph.IM↗

arXiv2022

Supernova Light Curves Approximation based on Neural Network Models

Photometric data-driven classification of supernovae becomes a challenge due to the appearance of real-time processing of big data in astronomy. Recent studies have demonstrated the superior quality of solutions based on various machine learning models. These models learn to classify supernova types using their light curves as inputs. Preprocessing these curves is a crucial step that significantly affects the final quality. In this talk, we study the application of multilayer perceptron (MLP), bayesian neural network (BNN), and normalizing flows (NF) to approximate observations for a single light curve. We use these approximations as inputs for supernovae classification models and demonstrate that the proposed methods outperform the state-of-the-art based on Gaussian processes applying to the Zwicky Transient Facility Bright Transient Survey light curves. MLP demonstrates similar quality as Gaussian processes and speed increase. Normalizing Flows exceeds Gaussian processes in terms of approximation quality as well.

astro-ph.IM↗

arXiv2021

Physical modeling of viscous disc evolution around magnetized neutron star. Aql X-1 2013 outburst decay

We present a model of a viscously evolving accretion disc around a magnetized neutron star. The model features the varying outer radius of the hot ionized part of the disc due to cooling and the varying inner radius of the disc due to interaction with the magnetosphere. It also includes hindering of accretion on the neutron star because of the centrifugal barrier and irradiation of the outer disc and companion star by X-rays from the neutron star and disc. When setting inner boundary conditions, we take into account that processes at the inner disc occur on a time scale much less than the viscous time scale of the whole disc. We consider three types of outflow from the disc inner edge: zero outflow, one based on MHD calculations, and a very efficient propeller mechanism. The light curves of an X-ray transient after the outburst peak can be calculated by a corresponding, publicly available code. We compare observed light curves of the 2013 burst of Aql X-1 in X-ray and optical bands with modeled ones. We find that the fast drop of the $0.3-10$ keV flux can be solely explained by a radial shrinking of the hot disc. At the same time, models with the neutron star magnetic field $>10^8$ G have better fits because the accretion efficiency behaviour emphasizes the 'knee' on the light curve. We also find that a plato emission can be produced by a disc-reservoir with stalled accretion.

astro-ph.HE↗

arXiv2021

El-CID: A filter for Gravitational-wave Electromagnetic Counterpart Identification

As gravitational-wave (GW) interferometers become more sensitive and probe ever more distant reaches, the number of detected binary neutron star mergers will increase. However, detecting more events farther away with GWs does not guarantee corresponding increase in the number of electromagnetic counterparts of these events. Current and upcoming wide-field surveys that participate in GW follow-up operations will have to contend with distinguishing the kilonova from the ever increasing number of transients they detect, many of which will be consistent with the GW sky-localization. We have developed a novel tool based on a temporal convolutional neural network architecture, trained on sparse early-time photometry and contextual information for Electromagnetic Counterpart Identification (El-CID). The overarching goal for El-CID is to slice through list of new transient candidates that are consistent with the GW sky localization, and determine which sources are consistent with kilonovae, allowing limited target-of-opportunity resources to be used judiciously. In addition to verifying the performance of our algorithm on an extensive testing sample, we validate it on AT2017gfo - the only EM counterpart of a binary neutron star merger discovered to date - and AT2019npv - a supernova that was initially suspected as a counterpart of the gravitational-wave event, GW190814, but was later ruled out after further analysis.

astro-ph.HE↗

arXiv2021

Photometric Data-driven Classification of Type Ia Supernovae in the Open Supernova Catalog

We propose a novel approach for a machine-learning-based detection of the type Ia supernovae using photometric information. Unlike other approaches, only real observation data is used during training. Despite being trained on a relatively small sample, the method shows good results on real data from the Open Supernovae Catalog. We also investigate model transfer from the PLAsTiCC simulations train dataset to real data application, and the reverse, and find the performance significantly decreases in both cases, highlighting the existing differences between simulated and real data.

astro-ph.IM↗

arXiv2019

Fips: an OpenGL based FITS viewer

FITS (Flexible Image Transport System) is a common format for astronomical data storage. It was first standardised in the early 1980s. Even though astronomical data is now processed mostly using software, visual data inspection by a human is still important during equipment or software commissioning and while observing. We present Fips, a cross-platform FITS file viewer open source software. To the best of our knowledge, it is for the first time that the image rendering algorithms are implemented mostly on GPU (graphics processing unit). We show that it is possible to implement a fully-capable FITS viewer using OpenGL interface. We also emphasise the advantages of using GPUs for efficient image handling.

astro-ph.IM↗

arXiv2016

A viscous-convective instability in laminar Keplerian thin discs

Using the anelastic approximation of linearized hydrodynamic equations, we investigate the development of axially symmetric small perturbations in thin Keplerian discs. Dispersion relation is found as a solution of general Sturm-Liouville eigenvalue problem for different values of relevant physical parameters (viscosity, heat conductivity, disc semithickness). The analysis reveals the appearance of overstable mode for Prandtl parameter higher than some critical value. These modes have a viscous-convective nature and can serve as a seed for turbulence in astrophysical discs even in the absence of magnetic fields.

astro-ph.HE↗

arXiv2015

Vertical Convection in Turbulent Accretion Disks and Light Curves of the A0620-00 1975 Outburst

We present a model of the non-stationary $α$-disk with account for the irradiation and the vertical convection in the outer accretion disk where hydrogen is partially ionized. We include the viscous energy generation in the mix-length convection equations in accretion disks. The optical and X-ray light curves of X-ray nova A0620-00 are investigated in terms of this model. The turbulent viscosity parameter of the accretion disk is estimated, $α= 0.5 ÷0.6$, which is necessary to explain the luminosity decay rate on the descending branch of the X-ray light curve for the A0620-00 1975 outburst. The secondary luminosity maximum on the light curves is explained by assuming an additional injection of matter into the accretion disk from the optical companion.

astro-ph.HE↗