检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Balewski, Jan Amankwah, Mercy G. Bethel, E. Wes Perciano, Talita Van Beeumen, Roel National Energy Research Scientific Computing Center Lawrence Berkeley National Laboratory BerkeleyCA94720 United States Department of Mathematics Applied Mathematics and Statistics Case Western Reserve University ClevelandOH44106 United States Computer Science Department San Francisco State University San FranciscoCA94132 United States Scientific Data Division Lawrence Berkeley National Laboratory BerkeleyCA94720 United States Applied Mathematics and Computational Research Division Lawrence Berkeley National Laboratory BerkeleyCA94720 United States

The novel constructive EHands protocol defines a universal set of quantum operations for multivariable polynomial transformations on quantum processors by introducing four basic subcircuits—multiplication, addition, negation, and parity flip—and by using expectation-value encoding (EVEN) to represent real numbers in quantum states. These elementary arithmetic operations can be systematically composed to compute degree-d polynomials, Pd(x), on a QPU. The resulting quantum circuit structure closely mirrors the stepwise evaluation of polynomials on a classical calculator, providing an intuitive and efficient approach to polynomial computation on quantum hardware. By enabling direct, predictable polynomial and nonlinear data transformations on a QPU, our method reduces dependence on classical post-processing in hybrid quantum-classical algorithms, enabling advancements in many quantum algorithms. The EHands quantum circuits are compact enough to deliver meaningful and accurate results on today’s noisy quantum processors. We present a detailed implementation of P4(x) and report experimental results for polynomial approximations of common functions, obtained using IBM’s Heron-class quantum processors and an ideal Qiskit simulator. Copyright © 2025, The Authors. All rights reserved.

关键词： Polynomial approximation

来源：评论

学校读者我要写书评

暂无评论

Feature enhancement model with up sampling based cyber threat attack detection and classification on imbalanced dataset in Industrial Internet of Things

引用

Alexandria Engineering Journal 2025年 128卷 247-258页

作者： Hayam Alamro Fahd N. Al-Wesabi Sultan Alahmari Jawhara Aljabri Shouki A. Ebad Asma Alshuhail Fouad Shoie Alallah Abdulrhman Alshareef Mahir Mohammed Sharif Department of Information Systems College of Computer and Information Sciences Princess Nourah bint Abdulrahman University P.O. Box 84428 Riyadh 11671 Saudi Arabia Department of Computer Science Applied College at Mahayil King Khalid University Saudi Arabia King Abdul Aziz City for Science and Technology (KACST) Cybersecurity Institute Riyadh Saudi Arabia Department of Computer Science University College in Umluj University of Tabuk Saudi Arabia Center for Scientific Research and Entrepreneurship Northern Border University Arar 73213 Saudi Arabia Department of Information Systems College of Computer Sciences & Information Technology King Faisal University Saudi Arabia Department of Information Systems Faculty of Computing and Information Technology King Abdulaziz University Jeddah Saudi Arabia Department of Computer and Self Development Preparatory Year Deanship Prince Sattam bin Abdulaziz University AlKharj Saudi Arabia

With the fast growth and utilization of artificial intelligence (AI) models, the Industrial Internet of Things (IIoT) has remarkably progressed in locating industrial communication and enhancing industrial methods quickly. In Industry 5.0, the hyper-automation method is a technical trend that navigates manufacturing objects to intellectual devices of IIoT, smart, agile software, cloud computing, smart robotics, and embedded modules by reliability and higher intricacy. Machine learning (ML) and deep learning (DL) methods were established for identifying anomalies by understanding the usual behaviour methods of the cyber threat attack and identifying and detecting deviations. This study proposes a Feature Enhancement Model with a White Shark Optimizer-based Cyber Threat Attack Detection and Classification (FEWSO-CTADC) technique on an Imbalanced Dataset in an IIoT environment. The primary focus of the FEWSO-CTADC technique is to enhance the automatic classification and detection of cyber threats in the IIoT environment. Initially, the FEWSO-CTADC technique implements a data preprocessing model to scale the raw information into a uniform format. Next, the SMOTE technique is used to manage the imbalanced class distribution in the attack recognition database. Moreover, the WSO-based feature subset selection is accomplished to select the superior set of features. Finally, the FEWSO-CTADC method utilizes the stacked auto-encoder (SAE) method for attack classification and recognition. Extensive experiments were conducted to demonstrate the improved performance of the FEWSO-CTADC approach, and the results were compared across various methods. The performance validation of the FEWSO-CTADC approach exhibited a superior value of 99.20 % over recent techniques under diverse metrics.

关键词： Cyber threat Stacked auto-encoder White shark optimizer Feature enhancement model Industrial Internet of Things

来源：评论

学校读者我要写书评

暂无评论

A Fast Rankability Method Based on Fuzzy Entropy

SSRN

引用

SSRN 2024年

作者： Kong, Lingping Velasquez, Juan D. Pant, Millie Pan, Jeng-Shyang Snášel, Václav Department of Computer Science VSB-Technical University of Ostrava Listpodu 17 Czech Republic Department of Industrial Engineering University of Chile O’Higgins 1058 Santiago Chile Department of Applied Mathematics and Scientific Computing Indian Institute of Technology Uttarakhand Roorkee247667 India College of Computer Science and Engineering Shandong University of Science and Technology China School of Artificial Intelligence Nanjing University of Information Science and Technology 219 Ningliu Road China Instituto Sistemas Complejos de Ingeniería Santiago Chile Mehta Family School of Data Science and Artificial Intelligence Indian Institute of Technology Uttarakhand Roorkee247667 India

Rankability is a fundamental concept that evaluates a dataset's inherent capacity to establish a meaningful ranking of its elements, with widespread applications in web search, data mining, cybersecurity, and machine *** its significance, relatively little focus has been on evaluating datasets for ranking suitability. Several rankability measures have emerged, such as comparing graphs representing datasets with complete dominance graphs using combinatorial and linear algebraic techniques. However, these approaches typically involve solving eigenvalue problems for large matrices or training regression models on extensive datasets. In this paper, we propose to adopt efficient and straightforward statistical metrics, entropy, and variance to develop a novel rankability method. Our method is designed to be fast and efficient, addressing the challenges posed by existing techniques. We conducted experiments using three football tournament datasets, and the results demonstrate a strong correlation scores (correlation coefficients) in terms of Spearman and Kendall's tau evaluations with Elo rating, a system developed by Arpad Elo in the 1950s and applied in many competitions. Specifically, the Spearman correlation between the proposed method and the Elo rating reached 0.9205 on Atlantic Coast data with a p-value of 0.0004, which is higher than the compared algorithm by 0.2424 in correlation score. © 2024, The Authors. All rights reserved.

关键词： Entropy

来源：评论

学校读者我要写书评

暂无评论

B-SPLINE CNNS ON LIE GROUPS 8

B-SPLINE CNNS ON LIE GROUPS

引用

8th International Conference on Learning Representations, ICLR 2020

作者： Bekkers, Erik J. Amsterdam Machine Learning Lab Informatics Institute University of Amsterdam Netherlands Centre for Analysis and Scientific Computing Applied Mathematics and Computer Science Eindhoven University of Technology Netherlands

Group convolutional neural networks (G-CNNs) can be used to improve classical CNNs by equipping them with the geometric structure of groups. Central in the success of G-CNNs is the lifting of feature maps to higher dimensional disentangled representations in which data characteristics are effectively learned, geometric data-augmentations are made obsolete, and predictable behavior under geometric transformations (equivariance) is guaranteed via group theory. Currently, however, the practical implementations of G-CNNs are limited to either discrete groups (that leave the grid intact) or continuous compact groups such as rotations (that enable the use of Fourier theory). In this paper we lift these limitations and propose a modular framework for the design and implementation of G-CNNs for arbitrary Lie groups. In our approach the differential structure of Lie groups is used to expand convolution kernels in a generic basis of B-splines that is defined on the Lie algebra. This leads to a flexible framework that enables localized, atrous, and deformable convolutions in G-CNNs by means of respectively localized, sparse and non-uniform B-spline expansions. The impact and potential of our approach is studied on two benchmark datasets: cancer detection in histopathology slides in which rotation equivariance plays a key role and facial landmark localization in which scale equivariance is important. In both cases, G-CNN architectures outperform their classical 2D counterparts and the added value of atrous and localized group convolutions is studied in detail. © 2020 8th International Conference on Learning Representations, ICLR 2020. All rights reserved.

关键词： Lie groups

来源：评论

学校读者我要写书评

暂无评论

Blockchain-Assisted Trust Framework for Increased Survivability in Internet of Vehicles

引用

Internet Technology Letters 2025年第3期8卷

作者： Sharma, Kanta Prasad Al-Rashed, Atheer Johar, M.G.M. Alsoud, Anas Ratib Singh, Amrita Dey, Protyay Pant, Shivani Amity University Uttar Pradesh Greater Noida India College of Administrative Sciences Applied Science University Al Eker Bahrain Management and Science University Selangor Shah Alam Malaysia Hourani Center for Applied Scientific Research Al-Ahliyya Amman University Amman Jordan Department of Computer Science and Engineering Chandigarh Engineering College Chandigarh Group of Colleges-Jhanjeri Mohali India NIMS School of Computing Science and Artificial Intelligence NIMS University Rajasthan Jaipur India Uttaranchal Institute of Technology Uttaranchal University Dehradun India

VANETs enhance traffic efficiency and road safety, but they can be attacked by malicious vehicles. These malicious vehicles can cause accidents or endanger lives by broadcasting false event messages and disrupting Internet of Vehicles applications. Before responding to sender messages, receiver vehicles must estimate the legitimacy and trustworthiness of the source vehicles. Existing solutions struggle to balance security and efficiency effectively. This paper introduces a model that combines the advantages of blockchain technology and trust model models to improve the trustworthiness, efficacy, and security of vehicular networks, alongside secure trust-based optimized routing. Extensive experiments demonstrate the security and efficiency of the proposed model. The proposed model is adaptable to diverse VANET scenarios, addressing all security and privacy needs more comprehensively than current trust schemes. Efficiency analysis and simulation results show that our proposed framework outperforms baseline models, highlighting its security, effectiveness, and robustness in enhancing IoV communication security. © 2025 John Wiley & Sons Ltd.

关键词： Highway accidents

来源：评论

学校读者我要写书评

暂无评论

The Silent Majority: Demystifying Memorization Effect in the Presence of Spurious Correlations

arXiv

引用

arXiv 2025年

作者： You, Chenyu Dai, Haocheng Min, Yifei Sekhon, Jasjeet S. Joshi, Sarang Duncan, James S. Department of Electrical & Computer Engineering Yale University New HavenCT United States Scientific Computing and Imaging Institute University of Utah Salt Lake CityUT United States Department of Statistics and Data Science Yale University New HavenCT United States Department of Biomedical Engineering Yale University New HavenCT United States Department of Applied Mathematics & Statistics Stony Brook University Stony BrookNY United States Department of Computer Science Stony Brook University Stony BrookNY United States

Machine learning models often rely on simple spurious features - patterns in training data that correlate with targets but are not causally related to them, like image backgrounds in foreground classification. This reliance typically leads to imbalanced test performance across minority and majority groups. In this work, we take a closer look at the fundamental cause of such imbalanced performance through the lens of memorization, which refers to the ability to predict accurately on atypical examples (minority groups) in the training set but failing in achieving the same accuracy in the testing set. This paper systematically shows the ubiquitous existence of spurious features in a small set of neurons within the network, providing the first-ever evidence that memorization may contribute to imbalanced group performance. Through three experimental sources of converging empirical evidence, we find the property of a small subset of neurons or channels in memorizing minority group information. Inspired by these findings, we articulate the hypothesis: the imbalanced group performance is a byproduct of "noisy" spurious memorization confined to a small set of neurons. To further substantiate this hypothesis, we show that eliminating these unnecessary spurious memorization patterns via a novel framework during training can significantly affect the model performance on minority groups. Our experimental results across various architectures and benchmarks offer new insights on how neural networks encode core and spurious knowledge, laying the groundwork for future research in demystifying robustness to spurious correlation. © 2025, CC BY.

关键词： Neurons

来源：评论

学校读者我要写书评

暂无评论

A Generalization of Multi-Source Fusion-Based Framework to Stock Selection

SSRN

引用

SSRN 2023年

作者： Snášel, Václav Velasquez Silva, Juan Domingo Pant, Millie Georgiou, Dimitrios Kong, Lingping Faculty of Electrical Engineering and Computer Science VŠB-Technical University of Ostrava Czech Republic Mehta Family School of Data Science and Artificial Intelligence Indian Institute of Technology Uttarakhand Roorkee247667 India Departament of Industrial Engineering University of Chile Santiago Chile Faculty of Electrical Engineering and Computer Science National Technical University of Athens Greece Chile Department of Applied Mathematics and Scientific Computing Indian Institute of Technology Uttarakhand Roorkee247667 India

Selecting outstanding tech stocks for investment is challenging. Specifically, studying the investment for academic purposes is not mature enough due to the disarray of the publications and the `over' informative experience from profit-making websites. Often some authors entitle stock price prediction subject as a stock selection problem;both prediction and selection are just sub-section of portfolio management. Moreover, stock websites provide tons of potential criteria showing their perspective evaluation to simulate and monitor the stock market professionally, which causes an extra problem for academic studies screening out the less informative criteria. The paper generalizes a novel framework with a user-interactive interface for stock selection problems based on multi-source data fusion and decision-level fusion to enhance reliability that is limited by the narrow criteria performance and to enhance the strength of a model overcoming the weakness of a single-performed model. This framework benefits the time-series prediction and decision-making study. Besides, we propose adopting dynamic time warping to assist a task-learning process by customizing a loss function that improves the accuracy of data prediction. The experiment shows that the proposed method reduces the prediction log error by 28\% in a best case of real-situation data and decreases the warping cost by 20\%. Finally, we illustrate the proposed framework by implementing it in a real-world stock data selection. The results are practical and effective, further justified through a detailed ablation study. © 2023, The Authors. All rights reserved.

关键词： Investments

来源：评论

学校读者我要写书评

暂无评论

ComPile: A Large IR Dataset from Production Sources

arXiv

引用

arXiv 2023年

作者： Grossman, Aiden Paehler, Ludger Parasyris, Konstantinos Ben-Nun, Tal Hegna, Jacob Moses, William S. Monsalve Diaz, Jose M. Trofin, Mircea Doerfert, Johannes University of California Davis United States School of Computation Information and Technology Technical University of Munich Germany Center for Applied Scientific Computing Lawrence Livermore National Laboratory United States University of Minnesota Twin Cities United States Department of Computer Science University of Illinois Urbana-Champaign United States Division of Mathematics and Computer Science Argonne National Laboratory United States Google Inc. United States

Code is increasingly becoming a core data modality of modern machine learning research impacting not only the way we write code with conversational agents like OpenAI’s ChatGPT, Google’s Bard, or Anthropic’s Claude, the way we translate code from one language into another, but also the compiler infrastructure underlying the language. While modeling approaches may vary and representations differ, the targeted tasks often remain the same within the individual classes of models. Yet, relying solely on the ability of modern models to extract information from unstructured code does not take advantage of 70 years of programming language and compiler development by not utilizing the structure inherent to programs in the data collection. This detracts from the performance of models working over a tokenized representation of input code and precludes the use of these models in the compiler itself. To work towards the first intermediate representation (IR) based models, we fully utilize the LLVM compiler infrastructure, shared by a number of languages, to generate a 1.4T Llama 2 token dataset of LLVM IR. We generated this dataset from programming languages built on the shared LLVM infrastructure, including Rust, Swift, Julia, and C/C++, by hooking into LLVM code generation either through the language’s package manager or the compiler directly to extract the dataset of intermediate representations from production grade programs. Statistical analysis proves the utility of our dataset not only for large language model training, but also for the introspection into the code generation process itself as well as for training of machine-learned compiler components. Copyright © 2023, The Authors. All rights reserved.

关键词： Large datasets

来源：评论

学校读者我要写书评

暂无评论

A PCA-based fuzzy tensor evaluation model for multiple-criteria group decision making

引用

applied Soft computing 2023年 132卷

作者： Singh, Meenu Pant, Millie Kong, Lingping Alijani, Zahra Snášel, Václav Department of Applied Mathematics and Scientific Computing Indian Institute of Technology Roorkee Uttarakhand247667 India Mehta Family School of Data Science and Artificial Intelligence Indian Institute of Technology Roorkee Uttarakhand247667 India Faculty of Electrical Engineering and Computer Science VŠB-TU Ostrava Czech Republic Institute for Research and Applications of Fuzzy Modeling University of Ostrava 30. dubna 22 701 03 Ostrava Czech Republic

Multiple-criteria group decision-making (MCGDM) problems mainly consist of multiple factors and multiple Decision Makers (DMs) or Users, for which dimension extension is necessary when considering all the entries of DMs together. Tensor, a generalized form of a matrix, displays a multi-way array item, which is the most suitable and practical way to represent high-dimensional data without losing any information. In this paper, we first reduce the dimension through Principal Component Analysis (PCA), which helps consider the most-informative criteria. Then, we reintroduced the tensor as a fuzzy-form tensor for the MCGDM problem because of user information uncertainty. We choose Interval-valued Neutrosophic Fuzzy numbers (IVNFNs) as the basis for the tensor form because of their ability to distinguish between truth, indeterminacy, and falsity in the data. Lastly, a Generalized Interval-valued Neutrosophic Fuzzy Weighted Geometric (GIVNFWG) operator is defined. Moreover, a generalized framework for fuzzy-form tensors for high-dimensional MCGDM problems is proposed. The feasibility and efficiency of this proposed process is illustrated for a real-world MCGDM problem of ranking the most efficient Third Party Reverse Logistics Partners (3PRLPs), i.e., recycled fiber-based paper mills for the packaging industry. The obtained results are according to the experts and are validated using sensitivity analysis. This analysis facilitates in assessing the impact on the overall ranking performance of 3PRLPs by considering various combinations of environment and technological sub-criteria. © 2022 Elsevier B.V.

关键词： Principal component analysis

来源：评论

学校读者我要写书评

暂无评论

Robusta: Robust AutoML for Feature Selection via Reinforcement Learning

arXiv

引用

arXiv 2021年

作者： Wang, Xiaoyang Li, Bo Zhang, Yibo Kailkhura, Bhavya Nahrstedt, Klara Department of Computer Science University of Illinois at Urbana-Champaign United States Center for Applied Scientific Computing Lawrence Livermore National Laboratory United States

Several AutoML approaches have been proposed to automate the machine learning (ML) process, such as searching for the ML model architectures and hyper-parameters. However, these AutoML pipelines only focus on improving the learning accuracy of benign samples while ignoring the ML model robustness under adversarial attacks. As ML systems are increasingly being used in a variety of mission-critical applications, improving the robustness of ML systems has become of utmost importance. In this paper, we propose the first robust AutoML framework, Robusta–based on reinforcement learning (RL)–to perform feature selection, aiming to select features that lead to both accurate and robust ML systems. We show that a variation of the 0-1 robust loss can be directly optimized via an RL-based combinatorial search in the feature selection scenario. In addition, we employ heuristics to accelerate the search procedure based on feature scoring metrics, which are mutual information scores, tree-based classifiers feature importance scores, F scores, and Integrated Gradient (IG) scores, as well as their combinations. We conduct extensive experiments and show that the proposed framework is able to improve the model robustness by up to 22% while maintaining competitive accuracy on benign samples compared with other feature selection methods. Copyright © 2021, The Authors. All rights reserved.

关键词： Feature extraction

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：