Searcharxiv⌕ Search

arXiv subjects

Sergei Popov

Publications and source records attributed to Sergei Popov.

At least 19 recordsLinked to original sources
arXiv2026

Population synthesis of low-mass binaries with wind-accreting neutron stars

We perform population synthesis modeling of neutron star binary systems containing a low-mass main-sequence companion underfilling its Roche lobe. About $10^5$ such systems are expected to exist in the Milky Way. In the majority of such systems, NSs are expected to start accreting the stellar wind matter within a few Gyr, resulting in the appearance of a dim X-ray source. The exact value of this fraction mainly depends on the efficiency of the propeller spin-down, properties of the stellar wind, and common envelope efficiency. Mostly, NSs in such systems are formed in electron capture supernovae. Thus, they obtain a low natal kick. Due to this, the center-of-mass velocities of such binaries are low, too. Only in the case of high common envelope efficiency ($α_\mathrm{CE}\gtrsim 3$), a few tens of long-lived high-velocity binaries can be produced in core-collapse supernovae.

astro-ph.HE↗
arXiv2026

Can accreting isolated neutron stars be detected?

We perform population synthesis modeling of isolated neutron stars in the Milky Way over its lifetime. Compared with previous studies, we use more detailed models of the interstellar medium and the magneto-rotational evolution of neutron stars. We demonstrate that presently, the spin-down rate at the propeller stage is the main uncertain factor that influences the number of accreting isolated neutron stars. If the propeller stage duration allows neutron stars to begin accreting matter from the interstellar medium and if the efficiency of accretion is high, then the number of accreting isolated neutron stars in eROSITA data can reach ~a few thousand. Still, uncertainties in spin-down at the propeller stage and in the accretion process can drastically decrease this number. We suggest that future observations of neutron stars in wide low-mass binaries recently discovered by Gaia can clarify these issues.

astro-ph.HE↗
arXiv2025

Natal kicks of compact objects

When compact objects - neutron stars and black holes - are formed in a supernova explosion, they may receive a high velocity at formation, which may reach or even exceed 1000 km s-1 for neutron stars and hundreds of km s-1 for black holes. The origin of the velocity kick is intimately related to supernova physics. A better understanding of kick properties from astronomical observations will shed light on the unsolved problems of these explosions, such as the exact conditions leading to exotic electron capture and ultra-stripped supernovae. Kick velocities are profoundly important in several areas of astrophysics. Being a result of supernova explosions, the kick velocity distribution must be explained in the framework of the supernova mechanism. The kick magnitudes and directions influence many topics related to binary systems, including the rate of compact object coalescences observable through gravitational waves. Moreover, knowledge of the kick velocity distribution is significant in predicting future observational results and their interpretation. For example, it is expected that the Roman space telescope will discover many microlensing events related to neutron stars and black holes; accurate estimates of the number of observable microlensing events require precise kinematic properties of these compact objects.

astro-ph.HE↗
arXiv2025

Analytical Model for Light Scattering in Transparent Composites

Transparent composites that combine optical transmittance with mechanical performance are increasingly important for applications in optical devices, sustainable building materials, and photonic engineering. However, predicting light scattering in such materials remains a challenge due to complex, multi-scale microstructural interactions. Here, we present a physically grounded and computationally efficient analytical model. It predicts angular light scattering in transparent composites based on the Average Interface Number (AIN), a single governing microstructural metric derived in this work from geometrical optics. The model accurately captures angular scattering behavior in both fiber- and particle-reinforced composites, as well as in transparent wood. We further introduce the Equivalent Average Interface Number (EAIN), combining AIN with refractive index mismatch into a unified parameter for fast haze prediction. Deep neural network (DNN) analyses confirm AIN as the dominant feature influencing optical scattering. The model predictions are supported by ray-tracing simulations and experimental trends from literature. Finally, we demonstrate the application of our model in fast image rendering simulations through transparent composites. This work provides a compact and practical toolbox for optical design and optimization of transparent structural materials.

physics.optics↗
arXiv2025

Evolution of neutron stars in wide eccentric low-mass binary systems

Precise astrometric measurement with Gaia satellite resulted in the discovery of tens of wide binary systems consisting of a Sun-like star and an invisible component. The latter can be a white dwarf, a neutron star, or a black hole. In this paper, we model magneto-rotational evolution of neutron stars in wide low-mass binaries accounting for the orbital eccentricity. We aim to calculate when neutron stars in such systems can start to accrete matter from the stellar wind of the companion. We show that the transition from the ejector to the propeller stage occurs earlier in more eccentric systems, thus increasing the time that neutron stars can spend accreting matter. Our calculations show that in the case of efficient spin-down at the propeller stage, a neutron star in an eccentric orbit with $e\gtrsim0.6$ and a standard magnetic field $B=10^{12}$ G can start accreting within a few Gyr. For neutron stars with $B=10^{13}$ G the onset of accretion occurs earlier regardless of the orbital eccentricity. Otherwise, with a lower spin-down rate, such a neutron star will remain at the propeller stage for most of its life.

astro-ph.HE↗
arXiv2023

Experimental Validation of Coherent Joint Transmission in a Distributed-MIMO System with Analog Fronthaul for 6G

The sixth-generation (6G) mobile networks must increase coverage and improve spectral efficiency, especially for cell-edge users. Distributed multiple-input multiple-output (D-MIMO) networks can fulfill these requirements provided that transmission/reception points (TRxPs) of the network can be synchronized with sub nanosecond precision, however, synchronization with current backhaul and fronthaul digital interfaces is challenging. For 6G new services and scenarios, analog radio-over-fiber (ARoF) is a prospective alternative for future mobile fronthaul where current solutions fall short to fulfill future demands on bandwidth, synchronization, and/or power consumption. This paper presents an experimental validation of coherent joint transmissions (CJTs) in a two TRxPs D-MIMO network where ARoF fronthaul links allow to meet the required level of synchronization. Results show that by means of CJT a combined diversity and power gain of +5 dB is realized in comparison with a single TRxP transmission.

eess.SP↗
arXiv2023

Symbolic expression generation via Variational Auto-Encoder

There are many problems in physics, biology, and other natural sciences in which symbolic regression can provide valuable insights and discover new laws of nature. A widespread Deep Neural Networks do not provide interpretable solutions. Meanwhile, symbolic expressions give us a clear relation between observations and the target variable. However, at the moment, there is no dominant solution for the symbolic regression task, and we aim to reduce this gap with our algorithm. In this work, we propose a novel deep learning framework for symbolic expression generation via variational autoencoder (VAE). In a nutshell, we suggest using a VAE to generate mathematical expressions, and our training strategy forces generated formulas to fit a given dataset. Our framework allows encoding apriori knowledge of the formulas into fast-check predicates that speed up the optimization process. We compare our method to modern symbolic regression benchmarks and show that our method outperforms the competitors under noisy conditions. The recovery rate of SEGVAE is 65% on the Ngyuen dataset with a noise level of 10%, which is better than the previously reported SOTA by 20%. We demonstrate that this value depends on the dataset and can be even higher.

cs.LG↗
arXiv2022

Search for gravitational-neutrino correlations on ground-based detectors

The problem of joint data processing from ground-based gravitational and neutrino detectors is considered in order to increase the detection efficiency of collapsing objects in the Galaxy. The development of the "neutrino - gravitational correlation" algorithm is carried out within the framework of the theory of optimal filtration as applied to the well-known OGRAN and BUST facilities located at the BNO INR RAS. The experience of analyzing neutrino and gravitational data obtained during the outburst of supernova SN1987A is used. Sequential steps of the algorithm are presented, formulas for estimating the statistical efficiency of a two-channel recorder are obtained.

gr-qc↗
arXiv2021

Photon-axion mixing in thermal emission of isolated neutron stars

Thermally emitting neutron stars represent a promising environment for probing the properties of axion-like particles. Due to the strong magnetic fields of these sources, surface photons may partially convert into such particles in the large magnetospheric region surrounding the stars, which will result in distinctive signatures in their spectra. However, the interaction depends on the polarization state of the radiation and is rather weak due to the low experimentally allowed values of the coupling constant $g_{γa}$. In this work, we compute the degree of photon-axion transition in the case of 100% O-mode polarization and spectral energy distribution of an isotropic blackbody with uniform surface temperature. The stellar magnetic field is assumed to be dipolar. We show that with the maximum effect reached for the magnetic fields $\sim10^{13}$ - $10^{14}$ G (typical for X-ray dim isolated neutron stars) and $g_{γa} = 2 \times 10^{-11}$ GeV$^{-1}$, the optical flux is reduced by 30 - 40%, while the high-energy part of the spectrum is not affected. The low-energy decrease exceeds 5% at $g_{γa} \geq 2 \times 10^{-12}$ GeV$^{-1}$ and $m_a \leq 2\times 10^{-6}$ eV, which is below the present experimental and astrophysical limits on axion parameters. To obtain the actual observational constraints, rigorous treatment of the radiative surface layers is required.

astro-ph.HE↗
arXiv2020

Embedding Words in Non-Vector Space with Unsupervised Graph Learning

It has become a de-facto standard to represent words as elements of a vector space (word2vec, GloVe). While this approach is convenient, it is unnatural for language: words form a graph with a latent hierarchical structure, and this structure has to be revealed and encoded by word embeddings. We introduce GraphGlove: unsupervised graph word representations which are learned end-to-end. In our setting, each word is a node in a weighted graph and the distance between words is the shortest path distance between the corresponding nodes. We adopt a recent method learning a representation of data in the form of a differentiable weighted graph and use it to modify the GloVe training algorithm. We show that our graph-based representations substantially outperform vector-based methods on word similarity and analogy tasks. Our analysis reveals that the structure of the learned graphs is hierarchical and similar to that of WordNet, the geometry is highly non-trivial and contains subgraphs with different local topology.

cs.CL↗
arXiv2020

Editable Neural Networks

These days deep neural networks are ubiquitously used in a wide range of tasks, from image classification and machine translation to face identification and self-driving cars. In many applications, a single model error can lead to devastating financial, reputational and even life-threatening consequences. Therefore, it is crucially important to correct model mistakes quickly as they appear. In this work, we investigate the problem of neural network editing $-$ how one can efficiently patch a mistake of the model on a particular sample, without influencing the model behavior on other samples. Namely, we propose Editable Training, a model-agnostic training technique that encourages fast editing of the trained model. We empirically demonstrate the effectiveness of this method on large-scale image classification and machine translation tasks.

cs.LG↗
arXiv2019

Compensation of tropospheric and ionospheric effects in gravitational sessions of the spacecraft RadioAstron

The possibility of compensating atmospheric influence in an experiment on precision measurement of gravitational redshift using the "RadioAstron" spacecraft (SC) is discussed. When a signal propagates from a ground-based tracking station to a spacecraft and back, interaction with the ionosphere and troposphere makes considerable contribution to the frequency shift. A brief overview of the physical effects determining this contribution is given, and the principles of calculation and compensation of the corresponding frequency distortions of radio signals are described. Then these approaches are used to reduce the atmospheric frequency shift of the "RadioAstron" spacecraft signal. The spacecraft hardware allows working in two communication modes: "one-way" and "two-way", in addition, two communication channels at different frequencies work simultaneously. "One-way" (SC - ground-based tracking station) communication mode, a signal is synchronized by the on board hydrogen frequency standard. The "two-way" (SC - ground-based tracking station - SC ) mode is synchronized by the ground hydrogen standard. The calculations performed allow us to compare the quality of compensation of atmospheric fluctuations performed by various methods and choose the optimal one.

astro-ph.IM↗
arXiv2019

Neural Oblivious Decision Ensembles for Deep Learning on Tabular Data

Nowadays, deep neural networks (DNNs) have become the main instrument for machine learning tasks within a wide range of domains, including vision, NLP, and speech. Meanwhile, in an important case of heterogenous tabular data, the advantage of DNNs over shallow counterparts remains questionable. In particular, there is no sufficient evidence that deep learning machinery allows constructing methods that outperform gradient boosting decision trees (GBDT), which are often the top choice for tabular problems. In this paper, we introduce Neural Oblivious Decision Ensembles (NODE), a new deep learning architecture, designed to work with any tabular data. In a nutshell, the proposed NODE architecture generalizes ensembles of oblivious decision trees, but benefits from both end-to-end gradient-based optimization and the power of multi-layer hierarchical representation learning. With an extensive experimental comparison to the leading GBDT packages on a large number of tabular datasets, we demonstrate the advantage of the proposed NODE architecture, which outperforms the competitors on most of the tasks. We open-source the PyTorch implementation of NODE and believe that it will become a universal framework for machine learning on tabular data.

cs.LG↗
arXiv2019

Reconfigurable frequency coding of triggered single photons in the telecom C--band

In this work, we demonstrate reconfigurable frequency manipulation of quantum states of light in the telecom C-band. Triggered single photons are encoded in a superposition state of three channels using sidebands up to 53 GHz created by an off-the-shelf phase modulator. The single photons are emitted by an InAs/GaAs quantum dot grown by metal-organic vapor-phase epitaxy within the transparency window of the backbone fiber optical network. A cross-correlation measurement of the sidebands demonstrates the preservation of the single photon nature; an important prerequisite for future quantum technology applications using the existing telecommunication fiber network.

cond-mat.mes-hall↗
arXiv2018

High-speed PAM4-based Optical SDM Interconnects with Directly Modulated Long-wavelength VCSEL

This paper reports the demonstration of high-speed PAM-4 transmission using a 1.5-μm single-mode vertical cavity surface emitting laser (SM-VCSEL) over multicore fiber with 7 cores over different distances. We have successfully generated up to 70 Gbaud 4-level pulse amplitude modulation (PAM-4) signals with a VCSEL in optical back-to-back, and transmitted 50 Gbaud PAM-4 signals over both 1-km dispersion-uncompensated and 10-km dispersion-compensated in each core, enabling a total data throughput of 700 Gbps over the 7-core fiber. Moreover, 56 Gbaud PAM-4 over 1-km has also been shown, whereby unfortunately not all cores provide the required 3.8 $\times$ 10 $^{-3}$ bit error rate (BER) for the 7% overhead-hard decision forward error correction (7% OH HDFEC). The limited bandwidth of the VCSEL and the adverse chromatic dispersion of the fiber are suppressed with pre-equalization based on accurate end-to-end channel characterizations. With a digital post-equalization, BER performance below the 7% OH-HDFEC limit is achieved over all cores. The demonstrated results show a great potential to realize high-capacity and compact short-reach optical interconnects for data centers.

eess.SP↗
arXiv2018

140 Gbaud On-Off Keying Links in C-Band for Short-Reach Optical Interconnects

We demonstrate 140 Gbaud intensity modulated direct detection dispersion-uncompensated links with Mach Zehnder modulator and distributed feedback travelling-wave electro-absorption modulator over 5500 and 960 meters of standard single mode fibre, respectively, enabled by compact packaged ultra-high speed InP-based 2:1-Selector.

eess.SP↗