• A
  • A
  • A
  • АБВ
  • АБВ
  • АБВ
  • A
  • A
  • A
  • A
  • A
Обычная версия сайта
  • RU
  • EN
  • HSE University
  • Publications
  • Book chapter
  • Нейросетевой метод поиска выбросов на основе анализа накопленной ошибки обучения
  • 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
June 5, 2026
Neural Network Maps as a Method for Constructing Mathematical Models
Scientists from HSE University–Nizhny Novgorod and the Institute of Physics Belgrade, Serbia, are jointly exploring the application of machine learning techniques and neural networks to the study of nonlinear dynamics. Natalya Stankevich, Leading Research Fellow at the Laboratory of Topological Methods in Dynamics of the Faculty of Informatics, Mathematics, and Computer Science at HSE University–Nizhny Novgorod, spoke to the HSE News Service about this international project.
June 5, 2026
‘In the Age of Technology, It Is Interesting to Look into the Past and Think about What We Can Take from It
Polina Tabakova decided to apply for a Philology degree at HSE in Nizhny Novgorod because she grew up in Mari El and did not want to move far away from the Russian forests. In an interview for the Young Scientists of HSE University project, she spoke about the genre of the campus novel, the existential drama of Kolobok, and a blackout version of Eugene Onegin.
June 5, 2026
HSE Scientists Develop Method to Compress Large Language Models Without Losing Quality
Researchers from the AI and Digital Science Institute at the HSE Faculty of Computer Science have developed a new compression method for large language models such as GPT and LLaMA that reduces their size by 25–36% without additional training or significant loss of accuracy. This is the first approach to use mathematical transformations—specifically, rotations of model weights—to make models more amenable to compression with structured matrices. The study results have been published in ACL Findings 2025. The code is available on GitHub.

 

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

?

Нейросетевой метод поиска выбросов на основе анализа накопленной ошибки обучения

.
Yasnitsky L., Голдобин М. А.
Language: Russian
Full text
Text on another site
Keywords: нейронная сетьвыбросОшибка обучениянакопленная ошибка

In book

Искусственный интеллект в решении актуальных социальных и экономических проблем ХХI века : Сборник статей по материалам Десятой всероссийской научно-практической конференции с международным участием (г. Пермь, ПГНИУ, 9–10 октября 2025 г.)
Пермский государственный национальный исследовательский университет, 2025.
Similar publications
Replacing Criterion of Creativity with Criterion of Investment for Results Created by Artificial Intelligence
Pakshin P., Legal Issues in the Digital Age 2026 Vol. 7 No. 1 P. 32–48
Artificial intelligence plays a significant role in automation, minimizing human intervention in fields such as medicine, art, and law. Despite the historically close relationship between art and technology, generative AI has expanded the potential for creative activity. A significant catalyst for this process has been the proliferation of pre-trained AI systems, which have accelerated the ...
Added: March 31, 2026
Нейронная сеть как инструмент поддержки принятия решений по ликвидации экономических последствий нештатных ситуаций
Мезенцев А. С., Ясницкий В. Л., Миролюбова Т. В. et al., В кн.: Искусственный интеллект в решении актуальных социальных и экономических проблем ХХI века : Сборник статей по материалам Десятой всероссийской научно-практической конференции с международным участием (г. Пермь, ПГНИУ, 9–10 октября 2025 г.).: Пермский государственный национальный исследовательский университет, 2025. С. 148–150.
На примере серийного производства стальных отливок показана возможность применения нейронных сетей в качестве системы поддержки принятия решений по снижению отрицательных экономических последствий некоторых нештатных ситуаций, связанных со сбоями поставок сырьевого материала. Ключевые слова: нештатная ситуация, производственный брак, нейронная сеть, прогнозирование, сырьевой материал, моделирование ...
Added: February 15, 2026
Искусственный интеллект в решении актуальных социальных и экономических проблем ХХI века : Сборник статей по материалам Десятой всероссийской научно-практической конференции с международным участием (г. Пермь, ПГНИУ, 9–10 октября 2025 г.)
Пермский государственный национальный исследовательский университет, 2025.
Представлены материалы Десятой всероссийской научно-практической конференции с международным участием «Искусственный интеллект в решении актуальных социальных и экономических проблем ХХI века», которая проводилась 9–10 октября 2025 г. в Перми, ПГНИУ. Сборник предназначен для научных и педагогических работников, преподавателей, аспирантов, магистрантов, студентов и всех, кто интересуется и занимается проблемами развития и применения методов искусственного интеллекта. ...
Added: February 15, 2026
Что изменилось во время ковид-кризиса в значимости ESG метрик в оценке снижения цен акций: поиск ответа с помощью объяснительного ИИ для российского фондового рынка
Teplova T., Sokolova T., Kissa D. et al., Журнал Новой экономической ассоциации 2026 Т. 70 № 1 С. 157–190
This study is the first attempt to apply Explainable Artifi cial Intelligence (ХAI) to reveal the relationship of different Environmental, Social and Governance (ESG) metrics of stock issuers on downside risk in the Russian market. The methodology is based on the two-stage approach, i. e., neural networks with dense layers and the Shapley values from ...
Added: October 17, 2025
Формирование требований к технологическим параметрам серийного производства на основе нейросетевого подхода
Yasnitsky L., Голдобин М. А., Прикладная информатика 2025 Т. 20 № 3(117) С. 85–100
Currently, artificial intelligence methods are widely used in the practice of serial production enterprises. They are used to detect defects, classify and eliminate them, identify the causes of defects, predict the quality and properties of the resulting product, select optimal parameters of the production process, and identify and study its patterns. However, outside the field ...
Added: July 10, 2025
Попытка проведения групповой экспертизы программных проектов с помощью генеративного искусственного интеллекта
Джейранян А. Д., Plaksin M. A., В кн.: Экономика 5.0: коллективный интеллект и развитие: материалы VIII Пермского экономического конгресса (г.Пермь, ПГНИУ, 1–2 февраля 2024 г.).: Пермь: ПГНИУ, 2024. С. 83–92.
The article evaluates the possibility of using currently publicly available generative artificial intelligence systems to organize group examination of software projects. The formulation of requests to chatbots (instructs, prompts) is proposed, which are designed to ensure that the necessary information is received. ...
Added: February 17, 2025
Нейросетевой алгоритм выявления и удаления выбросов в зашумленных наборах данных
Yasnitsky L., Plotnikova E. G., Прикладная информатика 2024 Т. 19 № 5 С. 88–100
Outliers in statistical data, which are the result of erroneously collected information, are often an obstacle to the successful application of machine learning methods in many subject areas. The presence of outliers in training data sets reduces the accuracy of machine learning models, and in some cases, makes the application of these methods impossible. Currently ...
Added: November 29, 2024
Parameter Estimation of Combinational Circuits Using Machine Learning Methods
A.A. Ziazetdinov, V.V. Zunin, I.I. Romanova, , in: 2024 International Russian Automation Conference (RusAutoCon).: IEEE, 2024. P. 937–942.
Added: September 19, 2024
Улучшенный алгоритм встраивания ЦВЗ в видео с возможностью извлечения с помощью камеры мобильного устройства
Evsyutin O., Melman A., Подболотов Д. И. et al., В кн.: Сборник трудов по материалам IX Международной конференции и молодежной школы "Информационные технологии и нанотехнологии (ИТНТ-2023)".: Самарский национальный исследовательский университет имени академика С.П. Королева, 2023.
Added: September 10, 2024
ПРИЛОЖЕНИЕ ПОИСКА, АНАЛИЗА И ПРОГНОЗИРОВАНИЯ ДАННЫХ В СОЦИАЛЬНЫХ СЕТЯХ
Slastnikov S., Zhukova L., Semichasnov I., Информационные технологии и вычислительные системы 2024 № 1 С. 97–108
In this article, we present a web service designed for searching, extracting, and analyzing data from social networks and messengers, demonstrating its application for studying communities within the "VKontakte" social network. The web service enables the identification of typical user profiles within communities, the assessment of emotional sentiment in posts and comments, as well as ...
Added: August 12, 2024
Об управлении химическим составом сырьевого материала и режимом плавки для обеспечения требуемых механических свойств стальных изделий серийного производства
Yasnitsky L., Мезенцев А. С., Прикладная математика и вопросы управления 2023 № 3 С. 109–126
A The goal of the work is to create a mathematical model suitable for operational control of the strength characteristics of the resulting steel product in the conditions of serial steelmaking. Existing approaches based on the results of testing prototypes obtained in laboratory conditions are not suitable for this purpose, since in the conditions of serial steelmaking, the ...
Added: February 7, 2024
Обзор методов стегоанализа с использованием нейронных сетей
Kosmachev A., Задорожникова А. А., Perov A., В кн.: БОЛЬШИЕ ДАННЫЕ Материалы I Международного форума (Новосибирск, 16–18 ноября 2022 года).: Новосибирск: Новосибирский государственный университет экономики и управления «НИНХ», 2023.
The article discusses the basic concepts and terms used in steganography, substantiates the relevance of the problem of steganalysis, discusses the use of deep neural networks in the tasks of steganalysis on digital images. A comparative analysis and description of the most effective convolutional network architectures for solving the task is performed. ...
Added: January 26, 2024
Нейросетевая система прогнозирования результатов сезона итальянской футбольной лиги "Серия А"
Бутина Д., Yasnitsky L., Вестник Пермского университета. Серия: Математика. Механика. Информатика 2023 № 1(60) С. 84–92
The article describes the development of neural network system for predicting the Italian football Lega Serie A season results. To select the initial set, thematic sites containing complete statistics on the necessary characteristics were used. The system based on cost characteristics has 12 input parameters. The average testing error of this system was 3 %. ...
Added: November 20, 2023
Прогнозирование распределения температуры по стволу нефтяных скважин при пароциклической обработке
Степанов В. А., Yasnitsky L., Экспозиция Нефть Газ 2023 № 3 С. 69–73
One of the main sources of obtaining information about the degree of heating of the oil reservoir and the effectiveness of steam cyclic treatment of wells is geophysical research, which consists in measuring the temperature in the wellbore using a descent geophysical instrument. This is a rather laborious and not always successful process. As an alternative, this ...
Added: November 20, 2023
Simulation of Residential Real Estate Markets in the Largest Russian Cities
Alekseev A., Economy of Regions 2022 Vol. 18 No. 2 P. 609–622
The existing mass appraisal models and mathematical tools for predicting the market value of residential property have a number of disadvantages, as they are developed for individual regions. Without considering the constantly changing economic environment, these models quickly become outdated and require constant updating. Thus, they are not suitable for construction business optimisation. The study ...
Added: November 19, 2023
Разработка метода и определение нового условия безопасности движения в состоянии многопоточности на основе математического моделирования
Zykov S. V., Золотухина М. А., Золотухин С. А., Модели, системы, сети в экономике, технике, природе и обществе 2023 № 3 С. 98–114
Background. All attention is focused on reducing the number of accidents on an unfavorable section of the Altufyevsky highway in the area of 5 intersections. Such incidents become aggravating circumstances for nearby infrastructure facilities, such as: shopping malls, shops, cafes, there are also metro stations on a busy site. The endless flow of people and cars, public transport stops, ...
Added: November 2, 2023
Finite-size analysis in neural network classification of critical phenomena
Chertenkov V., Burovskiy E., Shchur L., Physical Review E - Statistical, Nonlinear, and Soft Matter Physics 2023 Vol. 108 No. 3 Article L032102
We analyze the problem of supervised learning of ferromagnetic phas transitions from the statistical physics perspective. We consider two systems in two universality classes, the two-dimensional Ising model and two-dimensional Baxter-Wu model, and perform careful finite-size analysis of the results of the supervised learning of the phases of each model. We find that the variance ...
Added: September 19, 2023
Применение методов машинного обучения при автоматизации детектирования препятствия движению поезда через железнодорожный переезд
Искандеров Ю. М., Катарушкин Б. Е., Ершов А. А., Информатизация и связь 2020 № 2 С. 46–51
Aim. Currently, when creating intelligent information systems in various fields of practical activity, machine learning methods are used. The article shows the possibilities of using these methods in automating the detection of obstacles in the interest of improving safety and reducing the number of emergencies at level crossings. Materials and methods. The article discusses advanced ...
Added: September 15, 2023
Распознавание объектов по составляющим их примитивам и отношениям между ними
Сливницин П. А., Mylnikov L., Информатика и автоматизация (Труды СПИИРАН) 2023 Т. 22 № 3 С. 511–540
The paper’s goal is to develop a methodology and algorithm for the recognition of objects in the environment, keeping the quality with an increasing number of objects. For this purpose, the following problems were solved: recognition of the shape features, estimation of relations between features, and matching between the found features and relations and the ...
Added: May 24, 2023
Finite-size analysis in neural network classification of critical phenomena
Chertenkov V., Burovskiy E., Shchur L., / Series "stat-mech". 2023. No. 2305:0334.
We analyze the problem of supervised learning of ferromagnetic phase transitions from the statistical physics perspective. We consider two systems in two universality classes, the two-dimensional Ising model and two-dimensional Baxter-Wu model, and perform careful finite-size analysis of the results of the supervised learning of the phases of each model. We find that the variance ...
Added: May 8, 2023
Учебно-методические материалы мастер-класса «Состязательные атаки на нейронные сети распознавания изображений» для студентов и школьников
Pantiukhin D., Информатика и образование 2023 Т. 38 № 1 С. 55–63
The problem of neural network vulnerability has been the subject of scientific research and experiments for several years. Adversarial attacks are one of the ways to “trick” a neural network, to force it to make incorrect classification decisions. The very possibility of adversarial attack lies in the peculiarities of machine learning of neural networks. The ...
Added: April 14, 2023
Нейросетевая модель для определения регламентных параметров технологического процесса переработки рудного сырья
Мезенцев А. С., Yasnitsky L., Прикладная информатика 2022 Т. 17 № 6 С. 56–67
Machine learning methods are currently widely used to solve various production problems, the problems of defects diagnosing and predicting for items in mass production, in particular. One of the most important problems is defects diagnosing and predicting, basing on its solution the regulations for the technological processes parameters and raw materials used can be determined, that insures the minimum ...
Added: January 31, 2023
  • 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