DSA Faculty
API
← к списку преподавателей

Деркач Денис Александрович

Факультет компьютерных наук

Профиль на hse.ru ↗ тел.: 27266
Публикаций
226
Языков
3
Наград
6
Конференций
0
Профиль Публикации (226) Курсы (9)

Профессиональные интересы

физика высоких энергийанализ данных

Должности

  • Заведующий лабораторией — Факультет компьютерных наук, Институт искусственного интеллекта и цифровых наук, Научно-учебная лаборатория методов анализа больших данных
  • Доцент — Факультет компьютерных наук, Департамент больших данных и информационного поиска
  • Научный руководитель образовательной программы — Умные устройства: аппаратная разработка

Био

  • · Начал работать в НИУ ВШЭ в 2015 году.
  • · Научно-педагогический стаж: 7 лет.

Образование

  • 2010 · PhD: Университет Париж XI
  • 2007 · Магистратура: Санкт-Петербургский государственный университет, специальность «Физика», квалификация «Магистр»
  • 2004 · Бакалавриат: Санкт-Петербургский государственный университет, специальность «Физика», квалификация «Бакалавр»

Опыт работы

  • · 2017: н. в

Награды и поощрения

  • · Медаль "Признание - 10 лет успешной работы" НИУ ВШЭ (сентябрь 2025)
  • · Благодарность первого проректора НИУ ВШЭ (август 2024)
  • · Благодарность НИУ ВШЭ (май 2024)
  • · Почетная грамота Министерства науки и высшего образования Российской Федерации (ноябрь 2022)
  • · Благодарность первого проректора НИУ ВШЭ (август 2022)
  • · Благодарность Факультета компьютерных наук НИУ ВШЭ (август 2018)

Гранты и проекты

  • — · на соискание учёной степени кандидата наук

Идентификаторы исследователя

Публикации (226)

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

2023 · ARTICLE · en

Context. 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. Aims. 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. Methods. 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. Results. 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.

DOI ↗ PDF ↗

Neutron reconstruction in the BM@N experiment using machine learning

2023 · CHAPTER · en

At present, new compact highly granular neutron detector is being developed for the BM@N experiment. This detector will be used to identify neutrons, to measure their energies using time-of-flight method, neutron yields and azimuthal flow of neutrons in heavy-ion collisions at beam energies up to 4A GeV. Application of machine learning techniques and preliminary results of neutron identification and energy reconstruction are discussed. First predictions of the anisotropic flow of neutrons using state-of-the-art models of heavy-ion collisions are shown.

DOI ↗ PDF ↗

Precise determination of the 𝐵0s–𝐵⎯⎯⎯⎯0s oscillation frequency

2022 · ARTICLE · en

Mesons comprising a beauty quark and strange quark can oscillate between particle (𝐵0sBs0) and antiparticle (𝐵⎯⎯⎯⎯0sB¯s0) flavour eigenstates, with a frequency given by the mass difference between heavy and light mass eigenstates, Δms. Here we present a measurement of Δms using 𝐵0s→𝐷−sBs0→Ds−π+ decays produced in proton–proton collisions collected with the LHCb detector at the Large Hadron Collider. The oscillation frequency is found to be Δms = 17.7683 ± 0.0051 ± 0.0032 ps−1, where the first uncertainty is statistical and the second is systematic. This measurement improves on the current Δms precision by a factor of two. We combine this result with previous LHCb measurements to determine Δms = 17.7656 ± 0.0057 ps−1, which is the legacy measurement of the original LHCb detector.

DOI ↗ PDF ↗

Measurement of χc1(3872) production in proton-proton collisions at √s = 8 and 13 TeV

2022 · ARTICLE · en

The production cross-section of the χc1(3872) state relative to the ψ(2S) meson is measured using proton-proton collision data collected with the LHCb experiment at centre-of-mass energies of √s = 8 and 13TeV, corresponding to integrated luminosities of 2.0 and 5.4fb−1, respectively. The two mesons are reconstructed in the J/ψπ+π− final state. The ratios of the prompt and nonprompt χc1(3872) to ψ(2S) production cross-sections are measured as a function of transverse momentum, pT, and rapidity, y, of theχc1(3872) and ψ(2S) states, in the kinematic range 4

DOI ↗ PDF ↗

Measurement of the lifetimes of promptly produced Ωc0 and Ξc0 baryons

2022 · ARTICLE · en

A measurement of the lifetimes of the Ωc0 and Ξc0 baryons is reported using proton-proton collision data at a centre-of-mass energy of 13TeV, corresponding to an integrated luminosity of 5.4 fb−1 collected by the LHCb experiment. The Ωc0 and Ξc0 baryons are produced directly from proton interactions and reconstructed in the pK−K−π+ final state. The Ωc0 lifetime is measured to be 276.5 ± 13.4 ± 4.4 ± 0.7 fs, and the Ξc0 lifetime is measured to be 148.0 ± 2.3 ± 2.2 ± 0.2 fs, where the first uncertainty is statistical, the second systematic, and the third due to the uncertainty on the D0 lifetime. These results confirm previous LHCb measurements based on semileptonic beauty-hadron decays, which disagree with earlier results of a four times shorter Ωc0 lifetime, and provide the single most precise measurement of the Ωc0 lifetime.

DOI ↗ PDF ↗

Measurement of the W boson mass

2022 · ARTICLE · en

The W boson mass is measured using proton-proton collision data at s√s = 13 TeV corresponding to an integrated luminosity of 1.7 fb−1 recorded during 2016 by the LHCb experiment. With a simultaneous fit of the muon q/pT distribution of a sample of W → μν decays and the ϕ* distribution of a sample of Z → μμ decays the W boson mass is determined to be (formula) where uncertainties correspond to contributions from statistical, experimental systematic, theoretical and parton distribution function sources. This is an average of results based on three recent global parton distribution function sets. The measurement agrees well with the prediction of the global electroweak fit and with previous measurements.

DOI ↗

Artificial Intelligence for High Energy Physics

2022 · BOOK · en

The Higgs boson discovery at the Large Hadron Collider in 2012 relied on boosted decision trees. Since then, high energy physics (HEP) has applied modern machine learning (ML) techniques to all stages of the data analysis pipeline, from raw data processing to statistical analysis. The unique requirements of HEP data analysis, the availability of high-quality simulators, the complexity of the data structures (which rarely are image-like), the control of uncertainties expected from scientific measurements, and the exabyte-scale datasets require the development of HEP-specific ML techniques. While these developments proceed at full speed along many paths, the nineteen reviews in this book offer a self-contained, pedagogical introduction to ML models' real-life applications in HEP, written by some of the foremost experts in their area.

DOI ↗

Advances in Neural Computation, Machine Learning, and Cognitive Research VI : Selected Papers from the XXIV International Conference on Neuroinformatics, October 17-21, 2022, Moscow, Russia

2022 · BOOK · en

Methods of Stokes profile inversion based on spectral polarization analysis represent a powerful tool for obtaining information on magnetic and thermodynamic properties in the solar atmosphere. However, these methods involve solving the radiation transport equation. Over the past decades, several approaches have been developed to provide an analytical solution to the inverse problem, but despite its advantages, in many cases it requires large computing resources. Neural networks have been shown to be a good alternative to these methods, but in general they tend to be overly confident in their predictions. In this paper, the uncertainty estimation of atmospheric parameters prediction is presented. It is shown that deterministic networks containing partially-independent MLP blocks allow one to estimate uncertainty in predictions achieving the high accuracy results.

DOI ↗ PDF ↗

Observation of the Decay Λ0b→Λ+cτ−¯ντ

2022 · ARTICLE · en

The first observation of the semileptonic b-baryon decay Λ0b→Λ+cτ−¯ντ, with a significance of 6.1σ, is reported using a data sample corresponding to 3 fb−1 of integrated luminosity, collected by the LHCb experiment at center-of-mass energies of 7 and 8 TeV at the LHC. The τ− lepton is reconstructed in the hadronic decay to three charged pions. The ratio K=B(Λ0b→Λ+cτ−¯ντ)/B(Λ0b→Λ+cπ−π+π−) is measured to be 2.46±0.27±0.40, where the first uncertainty is statistical and the second systematic. The branching fraction B(Λ0b→Λ+cτ−¯ντ)=(1.50±0.16±0.25±0.23)% is obtained, where the third uncertainty is from the external branching fraction of the normalization channel Λ0b→Λ+cπ−π+π−. The ratio of semileptonic branching fractions R(Λ+c)≡B(Λ0b→Λ+cτ−¯ντ)/B(Λ0b→Λ+cμ−¯νμ) is derived to be 0.242±0.026±0.040±0.059, where the external branching fraction uncertainty from the channel Λ0b→Λ+cμ−¯νμ contributes to the last term. This result is in agreement with the standard model prediction.

DOI ↗ PDF ↗

Observation of the B0 →¯D*0 K+ π− and B0s →¯D*0 K−π+ decays

2022 · ARTICLE · en

The first observations of B0→¯D∗(2007)0K+π− and B0s→¯D∗(2007)0K−π+ decays are presented, and their branching fractions relative to that of the B0→¯D∗(2007)0π+π− decay are reported. These modes can potentially be used to investigate the spectroscopy of charm and charm-strange resonances and to determine the angle γ of the Cabibbo-Kobayashi-Maskawa unitarity triangle. It is also important to understand them as a source of potential background in determinations of γ from B+→DK+ and B0→DK+π− decays. The analysis is based on a sample corresponding to an integrated luminosity of 5.4 fb−1 of proton-proton collision data at 13 TeV center-of-mass energy recorded with the LHCb detector. The ¯D∗(2007)0 mesons are fully reconstructed in the ¯D0π0 and ¯D0γ channels with the ¯D0→K+π− decay. A novel weighting method is used to subtract background while simultaneously applying an event-by-event efficiency correction to account for resonant structures in the decays.

DOI ↗ PDF ↗

Курсы (9)