检索结果-内蒙古大学图书馆

Method with batching for stochastic finite-sum variational inequalities in non-Euclidean setting

CHAOS SOLITONS & FRACTALS 2024年 187卷

作者： Pichugin, Alexander Pechin, Maksim Beznosikov, Aleksandr Novitskii, Vasilii Gasnikov, Alexander Moscow Inst Phys & Technol Moscow Russia RAS Ivannikov Inst Syst Programming Moscow Russia Innopolis Univ Innopolis Russia

Variational inequalities are a universal optimization paradigm that incorporate classical minimization and saddle point problems. Nowadays more and more tasks require to consider stochastic formulations of optimization problems. In this paper, we present an analysis of a method that gives optimal convergence estimates for monotone stochastic finite-sum variational inequalities. In contrast to the previous works, our method supports batching, does not lose the oracle complexity optimality and uses an arbitrary Bregman distance to take into account geometry of the problem. Paper provides experimental confirmation to algorithm's effectiveness.

关键词： Stochastic optimization Variational inequalities Finite-sum problems Batching Bregman distance

来源：评论

学校读者我要写书评

暂无评论

Enhanced performance prediction of ATL model transformations

引用

PERFORMANCE EVALUATION 2024年 164卷

作者： Groner, Raffaela Bellmann, Peter Hoeppner, Stefan Thiam, Patrick Schwenker, Friedhelm Kestler, Hans A. Tichy, Matthias Ulm Univ Inst Software Engn & Programming Languages Albert Einstein Allee 11 D-89069 Ulm Baden Wurttembe Germany Ulm Univ Inst Neural Informat Proc Albert Einstein Allee 11 D-89069 Ulm Baden Wurttembe Germany Ulm Univ Inst Med Syst Biol Albert Einstein Allee 11 D-89069 Ulm Baden Wurttembe Germany

Model transformation languages are domain-specific languages used to define transformations of models. These transformations consist of the translation from one modeling formalism into another or just the updating of a given model. Such transformations are often described declaratively and are often implemented based on very small models that cover the language of the input model. As a result, transformation developers are often unable to assess the time required to transform a larger model. Hence, we propose a prediction approach based on machine learning which uses a set of model characteristics as input and provides a prediction of the execution time of a transformation defined in the Atlas Transformation Language (ATL). In our previous work (Groner et al., 2023), we already showed that support vector regression in combination with a model characterization based on the number of model elements, the number of references, and the number of attributes is the best choice in terms of usability and prediction accuracy for the transformations considered in our experiments. A major weakness of our previous approach is that it fails to predict the performance of transformations that also transform attribute values of arbitrary length, such as string values. Therefore, we investigate in this work whether an extension of our feature sets that describes the average size of string attributes can help to overcome this weakness. Our results show that the random forest approach in combination with model characterizations based on the number of model elements, the number of references, the number of attributes, and the average size of string attributes filtered by the 85th percentile of their variance is the best choice in terms of the simple way to describe a model and the quality of the obtained prediction. With this combination, we obtained a mean absolute percentage error (MAPE) of 5.07% over all modules and a MAPE of 4.82% over all modules excluding the transformation for wh

关键词： Performance prediction ATL Model transformation Linear regression Random forests Support vector regression

来源：评论

学校读者我要写书评

暂无评论

Digital Filter Architecture Based on Modified Winograd Method F(2x 2, 5x 5) and Residue Number system

引用

IEEE ACCESS 2023年 11卷 26807-26819页

作者： Valuev, Georgii Valueva, Maria Babenko, Mikhail Tchernykh, Andrei North Caucasus Fed Univ North Caucasus Ctr Math Res Stavropol 355017 Russia Ivannikov Inst Syst Programming Control Management & Appl Math Moscow 109004 Russia CICESE Res Ctr Comp Sci Dept Ensenada 22800 Mexico

Improving the characteristics of digital signal processing devices is an important task in many practical problems. The paper proposes the architecture of a two-dimensional digital filter with a $5\times 5$ mask, in which calculations are performed according to the Winograd method in the Residue Number system (RNS) with moduli of a special type. Theoretical analysis and hardware Field-Programmable Gate Array simulation are presented. The results show that the fragment throughput fr/s (number of fragments per second) of the device is 29.6%- 724.7% higher than state-of-the-art solutions. This is achieved by the combination of the Winograd method, which reduces the number of multiplications, with the RNS arithmetic, which performs addition and multiplication under smaller operands in parallel. However, our experiments showed that the proposed method requires up to 2.54%- 11.01% more Look-Up Tables and 3.58%- 19.83% higher power consumption compared to known analogues.

关键词： Filtering Finite impulse response filters Performance evaluation IIR filters Hardware Filtering algorithms Computer architecture Residue number system winograd method field-programmable gate array digital filter digital signal processing

来源：评论

学校读者我要写书评

暂无评论

Statistical Properties of the Galactic Magnetic Field in the Dynamo Model with Fluctuations of the Alpha Effect and Turbulent Diffusion

引用

BULLETIN OF THE LEBEDEV PHYSICS instITUTE 2024年第4期51卷 140-144页

作者： Grachev, D. A. Elistratov, S. A. Mikhailov, E. A. HSE Univ Moscow 109028 Russia Russian Acad Sci Ivannikov Inst Syst Programming Moscow 109004 Russia Shirshov Inst Oceanol Moscow 117997 Russia Russian Acad Sci Lebedev Phys Inst Moscow 119991 Russia Moscow MV Lomonosov State Univ Moscow 119991 Russia Skolkovo Inst Sci & Technol Moscow 121205 Russia

Magnetic fields in galaxies are usually originated through the action of a dynamo, which is based on the helicity of turbulent flows (alpha effect) and nonrigid rotation of a galaxy (differential rotation). They are opposed by dissipation, which tends to destroy the regular structures of the field. As a rule, dynamo equations contain averaged turbulence parameters. This approach presents difficulties in the case of galaxies with a highly inhomogeneous interstellar medium, associated, for example, with star formation. One way to solve the problem is to use random coefficients in the field equations. In this case, solutions require separate statistical analysis. This paper examines various statistical moments for the galactic magnetic field within the framework of a model assuming fluctuations of the alpha effect and turbulent diffusion.

关键词： magnetic field galaxies dynamo statistical moments

来源：评论

学校读者我要写书评

暂无评论

Revisiting linear machine learning through the perspective of inverse problems

引用

JOURNAL OF INVERSE AND ILL-POSED PROBLEMS 2025年第2期33卷 281-303页

作者： Liu, Shuang Kabanikhin, Sergey Strijhak, Sergei Wang, Ying-Ao Zhang, Ye Russian Acad Sci Ivannikov Inst Syst Programming Moscow Russia Novosibirsk State Univ Novosibirsk Russia Russian Acad Sci Siberian Branch Sobolev Inst Math Novosibirsk Russia Beijing Inst Technol Sch Math & Stat Beijing 100081 Peoples R China Shenzhen MSU BIT Univ MSU BIT SMBU Joint Res Ctr Appl Math Shenzhen 518172 Peoples R China

In this paper, we revisit Linear Neural Networks (LNNs) with single-output neurons performing linear operations. The study focuses on constructing an optimal regularized weight matrix Q from training pairs { G , H } {\{G,H\}} , reformulating the LNNs framework as matrix equations, and addressing it as a linear inverse problem. The ill-posedness of linear machine learning problems is analyzed through the lens of inverse problems. Furthermore, classical and modern regularization techniques from both the machine learning and inverse problems communities are reviewed. The effectiveness of LNNs is demonstrated through a real-world application in blood test classification, highlighting their practical value in solving real-life problems.

关键词： Machine learning linear neural network linear inverse and ill-posed problems regularization

来源：评论

学校读者我要写书评

暂无评论

Kotlin from the Point of View of Static Analysis Developer

引用

programming AND COMPUTER SOFTWARE 2023年第7期49卷 549-558页

作者： Afanasyev, V. O. Polyakov, S. A. Borodin, A. E. Belevantsev, A. A. Russian Acad Sci Ivannikov Inst Syst Programming Ul Solzhenitsyna 25 Moscow 109004 Russia Natl Res Univ Higher Sch Econ Ul Myasnitskaya 20 Moscow 101000 Russia Moscow MV Lomonosov State Univ Moscow 119991 Russia

This paper describes a static analysis tool for finding defects, analyzing metrics and relations for programs written in the Kotlin language. The approach is implemented in the Svace static analyzer developed at the Ivannikov institute for system programming of the Russian Academy of Sciences. The paper focuses on the problems we faced during the implementation, the approaches we used to solve them, and the experimental results for the tool we built. The tool not only supports Kotlin but is also capable of analyzing mixed projects that use both Java and Kotlin. We hope that this paper will be useful to static analysis developers and language designers.

关键词： static analysis defect search metric analysis vulnerabilities Kotlin JVM bytecode

来源：评论

学校读者我要写书评

暂无评论

Micro-CLK: Returning to the asynchronicity with Communication-Less Microkernel 21

Micro-CLK: Returning to the asynchronicity with Communicatio...

引用

12th ACM SIGOPS Asia-Pacific Workshop on systems (APSys)

作者： Klimiankou, Yauhen Russian Acad Sci Inst Syst Programming Moscow Russia

ISBN: (纸本)9781450386982

Inter-process communication (IPC) has always been the "Achilles heel" of microkernels, determining their overall performance. The entire history of microkernel development is tightly coupled to the debates about IPC, its efficiency, and the bottleneck it creates. The microkernel research community has consistently viewed the idea of message passing as the fundamental foundation of the microkernel, which is considered linked to its design inextricably. We are revising the place of IPC in the microkernel-based multi-server operating system design. We argue that communication can and potentially should be moved out to the userspace, making microkernel the first-tier resource manager solely. We describe a novel approach to out-of-kernel asynchronous messaging in multi-server microkernel OS and provide preliminary evaluation results to show its benefits.

关键词： asynchronous message passing multi-server operating system inter-process communication microkernel communication-less kernel

来源：评论

学校读者我要写书评

暂无评论

Ansible execution control in Python and Golang for cloud orchestration

引用

SOFTWAREX 2022年 19卷

作者： Badalyan, David Ivannikov, Oleg Borisenko RAS Ivannikov Inst Syst Programming Moscow Russia

The Ansible configuration manager is currently one of the most popular systems for software deployment. However, Ansible is difficult to debug when working with large scenarios and is difficult to embed into other systems. We propose the cotea (Python) and gocotea (Golang) tools that allow users to control Ansible execution programmatically, by iterating over the tasks and collecting progress information. We additionally propose gopython ??? a solution for embedding arbitrary Python code into a Golang application. The approach used was generalized up to the abstract architecture of the software tool, which allows for the control of an arbitrary Python program execution. ?? 2022 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license (http://***/licenses/by/4.0/).

关键词： Ansible Cloudorchestration Functionhooking Golang Python

来源：评论

学校读者我要写书评

暂无评论

Evaluating the complexity of deriving adaptive S'-homing and S′-synchronizing sequences for nondeterministic FSMs

引用

SOFTWARE QUALITY JOURNAL 2022年第1期30卷 161-180页

作者： Yevtushenko, Nina Kuliamin, Victor Kushik, Natalia Russian Acad Sci Ivannikov Inst Syst Programming 25 Alexander Solzhenitsyn St Moscow 109004 Russia Inst Polytech Paris Telecom SudParis CNRS SAMOVAR 9 Rue Charles Fourier F-91000 Evry France

Homing and synchronizing sequences (HSs and SSs) are used in Finite State Machine (FSM)-based testing for state identification when deriving a test suite with guaranteed fault coverage or when performing a non-intrusive (passive) testing or monitoring. However, such preset sequences do not always exist for nondeterministic FSMs or can be rather long if existing. Adaptive HSs and SSs are known to exist more often and can be much shorter which makes them attractive for deriving test suites and adaptive checking sequences. As nowadays, a number of specifications are represented by nondeterministic FSMs, there is a need for the deeper study of such sequences, their derivation strategies, and related complexity estimation / reduction. In this paper, given an FSM and a subset S' of its states, we are concerned with the existence check and derivation of S'-homing and S'-synchronizing adaptive sequences (test cases) which allow to reduce the uncertainty about the current state of the FSM up to a state of the subset S'. There are a number of research papers on evaluating the complexity of the existence check of adaptive HS / SS as well as on estimating the length of an HS / SS, and in the ICTSS'19 conference, we presented the complexity of their derivation for non-initialized complete nondeterministic FSMs. In this paper, we utilize the criteria of the ICTSS'19 paper to estimate the complexity of the existence check and derivation of S'-homing and S'-synchronizing test cases for non-initialized observable nondeterministic FSMs. Some sufficient conditions for weakly initialized FSMs to have adaptive S'-homing and S'-synchronizing test cases are also established.

关键词： Nondeterministic finite state machine (FSM) Adaptive S'-homing sequence Adaptive S'-synchronizing sequence

来源：评论

学校读者我要写书评

暂无评论

A Model-Based Approach to Automotive Feature Development for Updates and Upgrades 17

A Model-Based Approach to Automotive Feature Development for...

引用

17th IEEE International Congress on Intelligent and Service-Oriented systems Engineering (CISOSE)

作者： Schindewolf, Marc Wittler, Jan Willem Kuehn, Thomas Grimm, Daniel Sax, Eric Karlsruhe Inst Technol Inst Tech Informat Verarbeitung Karlsruhe Germany Karlsruhe Inst Technol Dependabil Software Intens Syst Karlsruhe Germany Martin Luther Univ Halle Wittenberg Software Engn & Programming Languages Halle Germany

ISBN: (纸本)9798350322392

The automotive industry faces challenges due to the increasing demand for customization and individualization by consumers. With a wide range of equipment levels and feature options, it has become impossible to test all possible configurations for a single vehicle during development and especially for vehicles already in service. As a basis for solving these challenges, integrated approaches for the development of upgradeable vehicles are required. In this context, future upgrade options should be considered and planned already in the design stage to enable sustainable business models for next vehicle generations. This paper proposes a refined variability model that combines two existing approaches to address the challenges faced by the industry. The same model can perform buildability or upgradeability checks and dynamic allocation optimization. The model is evaluated qualitatively by three exemplary use cases.

关键词： Automotive E/E architecture SOA software engineering reconfiguration model-driven software engineering software evolution software product lines

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：