• A
  • A
  • A
  • АБВ
  • АБВ
  • АБВ
  • A
  • A
  • A
  • A
  • A
Обычная версия сайта
  • RU
  • EN
  • HSE University
  • Publications
  • Articles
  • Coping with AI errors with provable guarantees
  • RU
  • EN
Расширенный поиск
Высшая школа экономики
Национальный исследовательский университет
Priority areas
  • business informatics
  • economics
  • engineering science
  • humanitarian
  • IT and mathematics
  • law
  • management
  • mathematics
  • sociology
  • state and public administration
by year
  • 2027
  • 2026
  • 2025
  • 2024
  • 2023
  • 2022
  • 2021
  • 2020
  • 2019
  • 2018
  • 2017
  • 2016
  • 2015
  • 2014
  • 2013
  • 2012
  • 2011
  • 2010
  • 2009
  • 2008
  • 2007
  • 2006
  • 2005
  • 2004
  • 2003
  • 2002
  • 2001
  • 2000
  • 1999
  • 1998
  • 1997
  • 1996
  • 1995
  • 1994
  • 1993
  • 1992
  • 1991
  • 1990
  • 1989
  • 1988
  • 1987
  • 1986
  • 1985
  • 1984
  • 1983
  • 1982
  • 1981
  • 1980
  • 1979
  • 1978
  • 1977
  • 1976
  • 1975
  • 1974
  • 1973
  • 1972
  • 1971
  • 1970
  • 1969
  • 1968
  • 1967
  • 1966
  • 1965
  • 1964
  • 1963
  • 1958
  • More
Subject
News
November 17, 2026
Physicists Show How a Solvent Shapes the Response of Drug Molecules
Many drug molecules can switch between different three-dimensional shapes. Using a mathematical model of ibuprofen dissolved in supercritical CO₂, researchers from HSE University and the Institute of Solution Chemistry of the Russian Academy of Sciences have shown that the solvent can make these molecules appear to respond unusually slowly. The molecules themselves continue to change shape rapidly, but their surroundings rearrange much more slowly, and this lag shows up in the measurements. The study has been published in The Journal of Chemical Physics. 
November 10, 2026
Physicists Train AI to Optimise Particle Beams in Accelerators
Researchers from the AI and Digital Science Institute at the HSE Faculty of Computer Science and the Budker Institute of Nuclear Physics at the Siberian Branch of the Russian Academy of Sciences have developed RLABC (Reinforcement Learning for Accelerator Beamline Control), a framework that uses reinforcement learning to identify optimal settings for a particle accelerator. In a computer simulation, RLABC optimised the accelerator's parameters to achieve a particle transmission rate of 70%. The study has been published in Computer Physics Communications.
November 9, 2026
'Organ-on-a-Chip Technology Could Eliminate the Need for Animal Testing'
The Laboratory of Molecular Physiology at the HSE Faculty of Biology and Biotechnology investigates how molecular regulators known as microRNAs control gene expression and contribute to the development of diseases, including cancer, viral and inflammatory conditions, and pregnancy complications. At HSE University, researchers develop organ-on-a-chip models: miniature devices containing living cells that mimic the functions of human organs, such as the placenta and intestines. These models help researchers understand how diseases develop and how drugs affect them. HSE News Service spoke with Evgeny Knyazev, Head of the Laboratory.

 

Have you spotted a typo?
Highlight it, click Ctrl+Enter and send us a message. Thank you for your help!

Publications
  • Books
  • Articles
  • Chapters of books
  • Working papers
  • Report a publication
  • Research at HSE

?

Coping with AI errors with provable guarantees

Information Sciences. 2024. Vol. 678. Article 120856.
Tyukin I., Tyukina T., van Helden D. P., Zheng, Z., Mirkes E. M., Sutton O. J., Zhou Q., Gorban A. N., Allison P.

AI errors pose a significant challenge, hindering real-world applications. This work introduces a novel approach to cope with AI errors using weakly supervised error correctors that guarantee a specific level of error reduction. Our correctors have low computational cost and can be used to decide whether to abstain from making an unsafe classification. We provide new upper and lower bounds on the probability of errors in the corrected system. In contrast to existing works, these bounds are distribution agnostic, non-asymptotic, and can be efficiently computed just using the corrector training data. They also can be used in settings with concept drifts when the observed frequencies of separate classes vary. The correctors can easily be updated, removed, or replaced in response to changes in distributions within each class without retraining the underlying classifier. The application of the approach is illustrated with two relevant challenging tasks: (i) an image classification problem with scarce training data, and (ii) moderating responses of large language models without retraining or otherwise fine-tuning.

Research target: Mathematics Computer Science
Language: English
Full text
DOI
Text on another site
Keywords: ошибки ИИИскусственный интеллектArtificial Intelligence (AI)Errors of AIAI error correctorsComputationally trustworthy AI systemsКорректоры ИИ ошибокДоверительные системы ИИ
Similar publications
Soliton turbulence within the modified Benjamin–Bona–Mahony equation
Flamarion M. V., Pelinovsky E., Physica D: Nonlinear Phenomena 2027 Vol. 500 Article 135452
In this work, we investigate soliton turbulence in the modified Benjamin–Bona–Mahony (mBBM) equation. A key advantage of this model is that it admits solitary wave solutions of both polarities while also being interpretable as a perturbation of the integrable modified Korteweg–de Vries (mKdV) equation. This framework enables a direct comparison of the main properties of soliton turbulence in ...
Added: November 18, 2026
Топология 3-многообразий, допускающих НМС-потоки с единственной седловой орбитой
Yagilev A., Pochinka O., Shubin D., Математический сборник 2026 Т. 217 № 11 С. 130–144
Рассматриваются неособые потоки Морса-Смейла (НМС-потоки) на замкнутых ориентируемых 3-многообразиях $M^3$, в предположении, что среди периодических орбит ровно одна седловая. Устанавливается, что  многообразие $M^3$ допускает такой поток  тогда и только тогда, когда оно гомеоморфно  линзовому пространству, связной сумме линзовых пространств или многообразию Зейферта вида $M(\mathbb S^2, (p_1,q_1), (p_2,q_2), (p_3,q_3))$. ...
Added: November 17, 2026
Active learning of collinear magnetic Moment Tensor Potentials using the spin-MLIP package from soft-constrained spin-polarized DFT calculations: a case study of Fe-Pd
Burov A., Котыхов А. С., Aksyonov D. et al., Computer Physics Communications 2027 Vol. 330 Article 110458
Explicit incorporation of magnetic degrees of freedom in machine-learning interatomic potentials (magnetic MLIPs) plays a crucial role in the correct description of magnetic materials and their properties. An important ingredient for fitting of magnetic MLIPs is spin-polarized density functional theory (DFT) calculations with non-equilibrium magnetic moments, i.e. DFT calculations with constraints on magnetic moments. In this study, we ...
Added: November 17, 2026
Toward reliable false discovery rate control in classification problems under distribution shift
Kertesz-Farkas A., Andrey B., Plos One 2026 Vol. 21 No. 11 P. 1–20
Background Domain shifts or batch effects can significantly degrade the performance of machine learning-based classifiers in software production without any warning. This can particularly be problematic in medical applications which involve human lives. The shift changes the shape of the prediction score distribution of the test samples compared to that of the training samples; hence, decision ...
Added: November 17, 2026
The medial septum model as a pacemaker of theta rhythm: the key role of short-term synaptic plasticity
Mysin I. E., Dubrovin S. V., Cognitive Neurodynamics 2026 Vol. 20 No. 191 P. 1–30
The hippocampal theta rhythm is crucial for memory and navigation. It is paced by parvalbumin-expressing GABAergic neurons in the medial septum. These neurons segregate into two subpopulations—Teevra and Komal cells—which fire in anti-phase bursts at theta frequency, with a stable phase shift of approximately 150 degrees, and exhibit mutual inhibition. While essential, the network mechanism ...
Added: November 14, 2026
Траектории ставок как сигнал сговора: LSTM-классификация аукционов государственных закупок
Ефимов К. Д., Журнал Новой экономической ассоциации 2026 № 72 С. 105–114
This study proposes a method for automatic detection of collusive behavior in electronic public procurement auctions based on analysis of the full bid trajectory. The empirical dataset comprises 508 auction histories from 2016–2020 collected from the RTS tender e-procurement platform; the positive class is labeled using case materials of the Federal Antimonopoly Service (FAS Russia), ...
Added: November 13, 2026
Model-Dependent Prompt-Component Ablation for Medical LLM Decision Support: Application to ENT Biologics Screening
Sergey N. Koltcov, Budkovaia M., Surkov A. et al., IEEE Journal of Biomedical and Health Informatics 2026 P. 1–10
Large language models (LLMs) show promise for clinical decision support, but privacy constraints and unreliable free-text decisions limit deployment. We present Medotor, a locally deployed system for screening patients with chronic rhinosinusitis with nasal polyps (CRSwNP) for biologic therapy, and evaluate prompt design on 163 retrospective cases (132 eligible and 31 ineligible). Five models were ...
Added: November 13, 2026
Upper Bound for Permanent Saturation of Metric Graphs using Interval Exchange Transformations
Ермолаев Е. С., Chernyshev V. L., Skripchenko A., Journal of Nonlinear Science 2026 Vol. 36 No. 122 (2026) P. 1–37
We study upper bounds for the moment of permanent -saturation in finite metric graphs. The dynamics is generated by moving points travelling with unit speed along edges and branching into all outgoing directions whenever they reach a vertex. We first reformulate this branched dynamics in terms of birth times at vertices and prove a sufficient ...
Added: November 12, 2026
Impact of 5G NR-V2X Message Losses on Connected Vehicle Warning Functions
Физулин А. В., Stepanyants V., Amerikanov A. et al., FUTURE TRANSPORTATION 2026 Vol. 6 No. 6 Article 264
Cooperative warning functions depend on whether V2X information becomes available in time to influence vehicle control. This article presents an event-based co-simulation framework coupling SUMO, ns-3 (ms-van3t), 5G-LENA NR sidelink and application records, with Sionna RT enabled for the intersection scenario. In 140 emergency-vehicle-warning runs, reducing mean packet reception ratio (PRR) from 0.9412 to 0.0675 ...
Added: November 12, 2026
On β-function of 𝒩 = 2 supersymmetric integrable sigma models. Part II
Alfimov Mikhail, Kurakin Andrey, Journal of High Energy Physics 2026 Vol. 2026 Article 99
We continue studying regularization scheme dependence of the 𝒩 = 2 supersymmetric sigma models. In the present work the previous result for the four loop β-function is extended to the five loop order. Namely, we find the renormalization scheme, in which the fifth loop contribution is completely eliminated, while the fourth loop contribution is represented ...
Added: November 12, 2026
Kripke Completeness For Predicate Modal Logics Of Bounded Alternativity: First Steps
Shehtman V. B., Shkatov D., Review of Symbolic Logic 2026 Vol. 19 No. 4 P. 859–882
We study Kripke completeness for predicate modal logics with propositional axioms of bounded alternativity corresponding to the condition on Kripke frames that every world sees at most n worlds. We prove that minimal predicate modal logics with axioms of bounded alternativity, as well as their extensions with the axiom of reflexivity, are strongly Kripke complete, whereas the ...
Added: November 12, 2026
On flexibility of trinomial varieties
Ignatev M., Vilkin T., Mediterranean Journal of Mathematics 2026 Vol. 23 No. 2 Article 70
Trinomial varieties are affine varieties given by a system of equations consisting of polynomials with three terms. Such varieties are total coordinate spaces of normal varieties with torus action of complexity one. For an affine variety X, we consider the subgroup SAut(X) of the automorphism group generated by all algebraic subgroups isomorphic to the additive ...
Added: November 12, 2026
Линейные отображения, сохраняющие матрицы с n или n − 1 различными собственными значениями
Maksaev A., Promyslov V., Тишутин П. С., Записки научных семинаров ПОМИ РАН 2026 Т. 554 С. 218–235
Let K be an algebraically closed field and let ℰₖ be the set of n × n matrices over K with exactly k distinct eigenvalues. Previously, the classification of linear maps T: Mₙ(K) → Mₙ(K) satisfying T(ℰₙ) ⊆ ℰₙ was obtained for invertible maps over fields of characteristic 0. In this paper, the classification is ...
Added: November 11, 2026
Nonexistence of Energy Function for Surface Diffeomorphisms with Nontrivial Basic Sets of Serpentine Geometric Type
Barinova M., Tirskaya Karina, Владикавказский математический журнал 2026 Vol. 28 No. 3 P. 25–43
In this paper, the problem of the existence of an energy function for discrete dynamical systems defined by structurally stable diffeomorphisms is studied. For the considered class of dynamical systems, such a function is defined as a continuous function that decreases along the orbits of the system outside the non-wandering set and remains constant on ...
Added: November 11, 2026
Автоматизированное построение математических теорий
Люксембург А. А., УРСС, 2005.
Изучается возможность автоматизированного построения математических теорий. Рассматривается дедуктивная система, основанная на языке логики предикатов первого порядка, объектами системы являются математические выражения или формулы, которые описывают математические объекты или их свойства. В дедуктивной системе выводятся математические определения и теоремы. Для доказательства теорем используются методы автоматического доказательства. Разработан алгоритм, выводящий часть формул системы. Для решения задачи используется аппарат математической ...
Added: October 7, 2026
Онтологический разрыв: когда Другой — алгоритм
Ugleva A. V., Социологическое обозрение 2026 Т. 25 № 3 С. 268–293
This article undertakes a socio-philosophical analysis of a situation in which the place of the Other within the structure of subjective experience is occupied by an algorithmic agent – an entity that simulates understanding yet remains devoid of genuine subjectivity. To investigate this configuration, the author turns to psychiatric practice as a representative case study: ...
Added: September 26, 2026
Пролетаризация труда в условиях развития искусственного интеллекта
Моисеев Г. В., Гуманитарные исследования в Восточной Сибири и на Дальнем Востоке 2026 № 2 (76) С. 58–68
The article presents a philosophical analysis of the transformation of labor in the context of the spread of AI technologies. Criticizing the notion of AI as an autonomous and neutral technology leading to the replacement of human labor, the author proposes to analyze the social consequences of AI development within the framework of automation processes ...
Added: September 17, 2026
Постатейный научно-практический комментарий к Федеральному закону от 26 июля 2026 г. № 243-ФЗ «О поддержке развития технологий искусственного интеллекта в Российской Федерации»
Белканов Е. А., Budnik R., Грекова Е. М. et al., М.: ООО "Сам полиграфист", 2026.
The commentary on Federal Law No. 243-FZ of July 26, 2026, "On Supporting the Development of Artificial Intelligence Technologies in the Russian Federation" was prepared by leading Russian scholars specializing in information (digital) law, including the legal aspects of artificial intelligence. It contains an article-by-article scientific and practical systematic analysis of the provisions of the ...
Added: September 5, 2026
Искусственный интеллект и этика будущего
Medushevsky A. N., Liberal.ru 2026
Today, more than ever before, the contribution of new technologies shapes the conditions of humanity's existence as a biological species, prompting a debate about its prospects for survival. This process of transformation—which spans every sphere of social and moral regulation, from family relations and child-rearing to the very meaning of life—is dividing public opinion between ...
Added: August 26, 2026
ОСОБЕННОСТИ ПРЕПОДАВАНИЯ КИТАЙСКОГО ЯЗЫКА В УСЛОВИЯХ ЦИФРОВИЗАЦИИ ОБРАЗОВАНИЯ: ДИАГНОСТИКА ПРОБЛЕМ И ТЕХНОЛОГИЧЕСКИЕ ПУТИ ИХ РЕШЕНИЯ
Игнатенко И. И., Danilina K., Педагогическое образование и наука 2026 № 1 С. 51–56
This article examines the specific pedagogical challenges inherent in teaching the Chinese language, which arise from its distinctive linguistic features: tonality, hieroglyphic writing system, context-dependent vocabulary, and analytical grammar. The study proposes a framework for overcoming these difficulties through the targeted application of modern digital technologies and analyzes the didactic potential of tools based on ...
Added: July 21, 2026
Event-Driven Platform for Machine Vision Component Integration with Operation Center
Gadzhimirzaev S., Хельвас А. В., 2023 3rd International Conference on Innovative Research in Applied Science, Engineering and Technology (IRASET) Mohammedia, Morocco 2023 P. 1–6
The article proposes the architecture for eventdriven Emergency Operation Center with Machine Vision Component. Sources of information are analyzed and approaches to machine vision events for tactical situations detection and estimation are discussed. Messages from Machine Vision Components are converted to Common Alerting Protocol and processed by Operation Center environment for tactical situations recognition. ...
Added: June 26, 2026
AI for good innovate for impact report 2025
Geneva: International Telecommunication Union, 2025.
Added: May 26, 2026
Социально-психологические факторы доверия искусственному интеллекту: состояние исследований
Samoilov O., Tatarko A., Вопросы теоретической экономики 2026 № 2 С. 209–228
The article presents a theoretical review of literature from the last ten years devoted to the analysis of socio-psychological factors of trust in artificial intelligence. The widespread adoption of automated artificial intelligence systems, which is associated with expected economic growth, reduced resource costs, and the optimization of various work processes, often faces user distrust in ...
Added: May 21, 2026
Human-in-the-Loop Egyptology: A System for Ancient Egyptian Text Study
Humonen I., Golyadkin M., Rubanova V. et al., , in: CHI EA '26: Proceedings of the Extended Abstracts of the 2026 CHI Conference on Human Factors in Computing Systems.: NY: Association for Computing Machinery (ACM), 2026. P. 1–5.
We present a prototype web-based human-in-the-loop system for studying Ancient Egyptian hieroglyphic texts that integrates an image-to-text pipeline into an interactive workspace for iterative refinement. In a formative user study with Egyptologists and students, the system enabled faster work and higher-quality results than a manual workflow. ...
Added: April 14, 2026
  • About
  • About
  • Key Figures & Facts
  • Sustainability at HSE University
  • Faculties & Departments
  • International Partnerships
  • Faculty & Staff
  • HSE Buildings
  • HSE University for Persons with Disabilities
  • Public Enquiries
  • Studies
  • Admissions
  • Programme Catalogue
  • Undergraduate
  • Graduate
  • Exchange Programmes
  • Summer University
  • Summer Schools
  • Semester in Moscow
  • Business Internship
  • Research
  • International Laboratories
  • Research Centres
  • Research Projects
  • Monitoring Studies
  • Conferences & Seminars
  • Academic Jobs
  • Yasin (April) International Academic Conference on Economic and Social Development
  • Media & Resources
  • Publications by staff
  • HSE Journals
  • Publishing House
  • iq.hse.ru: commentary by HSE experts
  • Library
  • Economic & Social Data Archive
  • Video
  • HSE Repository of Socio-Economic Information
  • HSE1993–2026
  • Contacts
  • Copyright
  • Privacy Policy
  • Site Map
Edit