检索结果-内蒙古大学图书馆

IEEE International Symposium on Information Theory

作者： Hsuan-Po Liu Mahdi Soleymani Hessam Mahdavifar Department of Electrical Engineering and Computer Science University of Michigan Ann Arbor MI USA Halıcıoğlu Data Science Institute University of California San Diego La Jolla CA USA

Distributed computing has attracted significant recent attention for speeding up large-scale computations by disseminating computational jobs from a central master node across several worker nodes/servers. However, worker nodes are often untrusted and can also collude to gain unauthorized access to sensitive data. Hence, sharing sensitive data with them raises data privacy concerns. Coded computing has emerged as a promising framework for speeding up distributed computing and can be also adapted to address security and privacy concerns utilizing tools from secret sharing and multi-party computing. However, ensuring perfect information-theoretic privacy imposes a strict threshold on the maximum number of colluding workers the protocol can tolerate and, also, necessitates quantizing/mapping data to finite fields. Differential privacy is a widely accepted practical measure to capture the privacy leakage of the shared data. The mainstream approach is then to add perturbations to the data via randomized mechanisms. In this paper, we revisit coded computing, and especially when it is adapted to handle real-valued data, and analyze the privacy guarantees through the lens of differential privacy in terms of the (ϵ,δ)-differential privacy metric, for the first time in the literature. All the computations are done over the field of real/complex numbers and data privacy, in terms of differential privacy, is attained by adding noise terms in a certain structured way. In particular, the noise is added through the secret sharing mechanism (which can be, in principle, decoded and cancelled out at the master) as means of ensuring differential privacy. Furthermore, we propose a differentially private distributed matrix multiplication protocol for matrix multiplications that keeps the privacy of data in the worst adversarial case.

关键词：

来源：评论

学校读者我要写书评

暂无评论

E-commerce Product Review Analysis based on Multi-class Support Vector Machine

E-commerce Product Review Analysis based on Multi-class Supp...

引用

Ambient Intelligence, Knowledge Informatics and Industrial Electronics (AIKIIE), International Conference on

作者： Laith H. Alzubaidi Ibraheem Hatem Mohammed Gotte Ranjith kumar B Sarada N. Radha The Islamic University Najaf Iraq Department of Computer Communications Engineering Al Rafidain University College Bagdad Iraq School of Computer Science & Artificial Intelligence SR University Warangal India Department of CSE-Data Science Malla Reddy Engineering College for women Hyderabad India Department of CSE Annapoorana Engineering College (Autonomous) Salem India

Recently, an extensive number of consumers can choose their best options from automated retailers and compare products in online stores. Sentiment analysis is frequently employed in applications meant to demonstrate and provide client service as the voice of the user. On E-commerce websites, user reviews provide helpful information about the product. Sentiment analysis of the text reviews helps forecast product sales by assessing user sentiment regarding the product. The Word Embedding Attention approach assigns additional weight to words that have a strong association with a specific class. Sentiment analysis machine learning models provide a thorough depiction of capabilities and far surpass conventional feature-based methods. The obj ective of this research is to improve sentiment analysis performance by creating a weighted ensemble with a Multi-class Support Vector Machine (MSVM) model utilising novel word embedding techniques. Because of the weighted ensemble's higher generalisation skills, MSVM produces superior results. When comparing existing methods for sentiment analysis, the Weighted ensemble with MSVM has produced higher results in terms of accuracy, precision, f-score, and recall. The Weighted ensemble with an MSVM has achieved 99.87 % accuracy, 99.39% precision, 99.03% f-score, and 99.690/0 recall in sentiment analysis when comparing existing methods such as LSTM+FL, APSO-LSTM, ABCDM, SSentiA, and Hybrid CNN+LSTM.

关键词：

来源：评论

学校读者我要写书评

暂无评论

General Strong Polarization

General Strong Polarization

引用

作者： Blasiok, Jaroslaw Guruswami, Venkatesan Nakkiran, Preetum Rudra, Atri Sudan, Madhu Department of Computer Science Columbia University 500 West 120th Street New YorkNY10027 United States Computer Science Department Carnegie Mellon University 5000 Forbes Avenue PittsburghPA15213 United States Halicioǧlu Data Science Institute University of California San Diego 10100 Hopkins Dr La Jolla CA92093 United States Department of Computer Science and Engineering University at Buffalo Davis 338 BuffaloNY14260 United States Harvard John A. Paulson School of Engineering and Applied Sciences Harvard University 33 Oxford Street CambridgeMA02138 United States

Arlkan's exciting discovery of polar codes has provided an altogether new way to efficiently achieve Shannon capacity. Given a (constant-sized) invertible matrix M, a family of polar codes can be associated with this matrix and its ability to approach capacity follows from the polarization of an associated [0, 1]-bounded martingale, namely its convergence in the limit to either 0 or 1 with probability 1. Arlkan showed appropriate polarization of the martingale associated with the matrix (G2 = (10/11) to get capacity achieving codes. His analysis was later extended to all matrices M that satisfy an obvious necessary condition for polarization. While Arlkan's theorem does not guarantee that the codes achieve capacity at small blocklengths (specifically in length, which is a polynomial in ( 1ϵ ) where (ϵ) is the difference between the capacity of a channel and the rate of the code), it turns out that a "strong"analysis of the polarization of the underlying martingale would lead to such constructions. Indeed for the martingale associated with (G2) such a strong polarization was shown in two independent works (Guruswami and Xia (IEEE IT'15) and Hassani et al. (IEEE IT'14)), thereby resolving a major theoretical challenge associated with the efficient attainment of Shannon capacity. In this work we extend the result above to cover martingales associated with all matrices that satisfy the necessary condition for (weak) polarization. In addition to being vastly more general, our proofs of strong polarization are (in our view) also much simpler and modular. Key to our proof is a notion of local polarization that only depends on the evolution of the martingale in a single time step. We show that local polarization always implies strong polarization. We then apply relatively simple reasoning about conditional entropies to prove local polarization in very general settings. Specifically, our result shows strong polarization over all prime fields and leads to efficient capacity-a

关键词： Polarization

来源：评论

学校读者我要写书评

暂无评论

Kernel Methods and Multi-layer Perceptrons Learn Linear Models in High Dimensions

arXiv

引用

arXiv 2022年

作者： Sahraee-Ardakan, Mojtaba Emami, Melikasadat Pandit, Parthe Rangan, Sundeep Fletcher, Alyson K. Department of Electrical and Computer Engineering UCLA Department of Statistics UCLA Halıcıoğlu Data Science Institute UCSD Department of Electrical and Computer Engineering NYU

Empirical observation of high dimensional phenomena, such as the double descent behaviour, has attracted a lot of interest in understanding classical techniques such as kernel methods, and their implications to explain generalization properties of neural networks. Many recent works analyze such models in a certain high-dimensional regime where the covariates are independent and the number of samples and the number of covariates grow at a fixed ratio (i.e. proportional asymptotics). In this work we show that for a large class of kernels, including the neural tangent kernel of fully connected networks, kernel methods can only perform as well as linear models in this regime. More surprisingly, when the data is generated by a kernel model where the relationship between input and the response could be very nonlinear, we show that linear models are in fact optimal, i.e. linear models achieve the minimum risk among all models, linear or nonlinear. These results suggest that more complex models for the data other than independent features are needed for high-dimensional analysis. Copyright © 2022, The Authors. All rights reserved.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Multi-Stage Vehicle Dispatch for Community Group-buying Logistics via Deep Reinforcement Learning

Multi-Stage Vehicle Dispatch for Community Group-buying Logi...

引用

International Conference on Mobile Ad-hoc and Sensor Networks, MSN

作者： Lingyi Xu Xingyuan Liang Xiaolei Zhou Tian He School of Computer Science and Engineering Southeast University The Sixty-Third Research Institute National University of Defense Technology Department of Technology and Data Intelligence JD Logistics

ISBN: (数字)9798350358261

ISBN: (纸本)9798350358278

Community group-buying is an emerging shopping model in which individuals within the same community collectively purchase daily necessities at lower prices. The logistics transportation in this scenario requires higher turnover rates and timeliness compared to traditional online shopping. Therefore, making efficient vehicle dispatch decisions is necessary to improve delivery efficiency and reduce transportation costs. However, uncertainties in orders and complex dependencies between multi-stage decisions in the community group-buying scenario challenge the scheduling of delivery vehicles. In this paper, we propose a vehicle dispatch system for community group-buying logistics based on a predict-then-optimize process, which incorporates a demand prediction module and a vehicle dispatch module. In the prediction module, we categorize goods and predict the total quantity of each category using a multi-layer perceptron (MLP). In the vehicle dispatch module, we model the vehicle dispatch process as a reinforcement learning problem and use an action masking mechanism to prune the policy search space. Experimental results demonstrate that the vehicle dispatch system achieves state-of-the-art performance compared to several popular baselines across three different datasets. Our system reduces transportation costs by at least 8.89% and improves decision efficiency by at least 10%. Moreover, it behaves close to the approximate optimal solution in most cases.

关键词： Space vehicles Costs Uncertainty Transportation Predictive models Search problems Sensors

来源：评论

学校读者我要写书评

暂无评论

Responsible AI on National Cybersecurity Strategy with Enhancement of Cybersecurity in Education

Responsible AI on National Cybersecurity Strategy with Enhan...

引用

Intelligent & Innovative Practices in engineering & Management (IIPEM), International Conference on

作者： Bhagyashree Ashok Tingare G. Prasanna Lakshmi Vijayshri Khedkar Rakesh Thoppaen Suresh Babu Tarun Dhar Diwan Pravin B Khatkale Department of Artificial Intelligence and Data Science D Y Patil College of Engineering Akurdi Pune School of Computer Science and Engineering Sandip University Nashik Maharashtra Department of Computer Science & Engineering Symbiosis Institute of Technology Symbiosis International (Deemed University) Pune India Senior Technical Architect HEXAWARE TECHNOLOGIES INC New Jersey Assistant Professor and Controller of Examination (COE) Atal Bihari Vajpayee University Bilaspur India Sanjivani University Kopargaon Maharashtra India

ISBN: (数字)9798350390049

ISBN: (纸本)9798350390056

For national cybersecurity to keep up with the digital transition, AI is more important than ever. Responsible AI improves cybersecurity and education by emphasizing openness and ethics. Ethical design is stressed from the outset in responsible AI. This article analyzes how artificial intelligence (AI) influences the national cybersecurity strategy, focusing on cybersecurity education development and effectiveness. The National Cybersecurity Strategy will frame this article. The major results show that responsible AI improves threat detection and response, cybersecurity infrastructure, talent shortage filling, collaborative research, and public trust. Educational institutions may now offer innovative training programs thanks to artificial intelligence. Completing this aim requires closing the gap between conventional methodologies and modern challenges. Closing gaps is vital. Functional artificial intelligence improves national cybersecurity and educational quality. This analysis shows that this conclusion is possible. This allows experts to properly react to shifting dangers. This is why it matters.

关键词： Training Ethics Focusing Collaboration Threat assessment Filling Artificial intelligence computer security

来源：评论

学校读者我要写书评

暂无评论

Skin cancer multiclass classification through different types of Convolutional Neural Networks 2

Skin cancer multiclass classification through different type...

引用

2nd International Conference on Computational Methods in science and Technology, ICCMST 2021

作者： Kumar, Mohit Gangadharan, Syam MacHinathu Parambil University Institute of Engineering Chandigarh University Department of Computer Science and Engineering Punjab Mohali India Sr Big Data Engineer General Mills Minnesota United States

ISBN: (纸本)9781665458665

In the human body, one of the major diseases found in some people across the globe is cancer. Cancer can be at any portion of the body like the breast, lungs, prostate and many more. One of the major difficulties during the analysis of cancer is the prediction of the type of cancer. This paper also discusses one of cancer, i.e., skin cancer, which are mainly seven types. To perform this multiclass classification, the work has done through the deep learning models based on Convolutional Neural Networks. Convolutional Neural Networks are implemented using pretrained model i.e., VGG-16. Firstly, segmentation is done with encoding and decoding using masking approach. Secondly, the model is evaluated and finally the prediction of different skin images for the cancer detection. Thereafter classification is done through Convolutional neural network and finally the model validation is done K cross fold method. © 2021 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Properties and applications of octonion fractional Fourier transform for 3-D octonion signals

引用

Digital Signal Processing: A Review Journal 2025年 165卷

作者： Yang, Xi Feng, Qiang Jiang, Nan Bhat, Mohammad Younus Urynbassarova, Didar School of Mathematics and Computer Science Yanan University Shanxi Yanan716000 China Shaanxi Key Laboratory of Intelligent Processing for Big Energy Data Shanxi Yanan716000 China Department of Mathematical Sciences Islamic University of Science and Technology Kashmir 192122 India National Engineering Academy of the Republic of Kazakhstan Almaty050000 Kazakhstan

In the field of signal processing, especially when dealing with complex signals and multidimensional data analysis requirements, traditional transformation methods often struggle to meet practical application needs. Although the Fourier transform and its derivative methods (such as the fractional Fourier transform) play an important role in signal processing, these methods have obvious limitations in handling high-dimensional non-stationary signals, particularly in capturing the intrinsic structure and time-frequency characteristics of multidimensional signals. To address this issue, this paper proposes a new signal processing tool: the octonion fractional Fourier transform (OFRFT). By systematically studying the differential properties of OFRFT and establishing its convolution theorem, the application of this method in multi-dimensional linear time-invariant systems, including series, parallel, and feedback connections, is thoroughly discussed. Simulation experiments based on octonion signals further verify the practical application value of OFRFT in signal processing. The results demonstrate that OFRFT not only provides a novel theoretical tool for analyzing the time-frequency characteristics of complex signals but also shows significant advantages in non-stationary signal processing, laying an important foundation for theoretical research and engineering applications in related fields. © 2025 Elsevier Inc.

关键词： Fourier series

来源：评论

学校读者我要写书评

暂无评论

Histopatolojik Görüntülerde Dogru Mitoz Tespiti için Geliştirilmiş Renk Normalleştirme Yöntemi: Enhanced Stain Normalization Method for Accurate Mitosis Detection in Histopathological Images

Histopatolojik Görüntülerde Dogru Mitoz Tespiti için Gel...

引用

International Conference on computer science and engineering (UBMK)

作者： Refik Samet Nooshin Nemati Emrah Hançer Serpil Sak Bilge Ayça Kırmızı Department of Computer Engineering Ankara University Ankara Türkiye Victoria University of Wellington Center for Data Science and Artificial Intelligence Wellington New Zealand Department of Medical Pathology Ankara University Ankara Türkiye

ISBN: (数字)9798350365887

ISBN: (纸本)9798350365894

Mitosis on H&E-stained (Hematoxylin and Eosin) images is an important prognostic marker for evaluating the tumor's aggressiveness and providing comprehensive and reliable information for accurate diagnosis and treatment. The variability in dosage rate and different fading rates in H&E stains lead to challenges in histopathological analysis. Standardizing the staining process and understanding the factors that influence stain stability is critical for consistency and accuracy in diagnosis. This paper proposes an improved stain normalization method with DDPM-based (Denoising Diffusion Probabilistic Models) diffusion model to detect mitosis with high accuracy in histopathological images. One of the challenges in the DDPM-based diffusion model is incorrect transfer, where the model may confuse hematoxylin with eosin due to the high diversity nature of DDPM-based diffusion models. A stain separation method using CD-SNMF (Color Deconvolution-Sparse Non- Negative Matrix Factorization) is proposed to address this issue. This method normalizes the mitosis in images in CCMCT (Canine Mast Cell Tumors) and MIDOG21 (Mitosis Domain Generalization Challenge 2021) datasets by this stain normalization pathway with decent performance.

关键词： Measurement Accuracy Histopathology Image color analysis Computational modeling Noise reduction Diffusion models Stability analysis Reliability Tumors

来源：评论

学校读者我要写书评

暂无评论

Identifying Influencing Factors for data Transactions:A Case Study from Shanghai data Exchange

引用

Journal of Systems science and Systems engineering 2020年第6期29卷 697-708页

作者： Qifeng Tang Zhiqing Shao Lihua Huang Wenyi Yin Yifan Dou Department of Computer Science and Engineering East China University of Science and TechnologyShanghai 200237China Shanghai Data Exchange Corporation Shanghai 200436China School of Management Fudan UniversityShanghai 200433China

In the age of artificial intelligence,firms'internal data are increasingly valuable when merged with each other for inter-firm analysis and ***,the inter-firm data transactions represent a novel challenge on pricing due to the complex nature of data,such as quality information asymmetry,lack of pricing standards,and the negligible marginal *** paper conducts a case study at Shanghai data Exchange to explore the factors that can facilitate the data transactions between buyers and *** use interview transcripts from 18 participating firms to construct our three theoretical dimensions:increasing the perceived value,mitigating the cost,and improving the market *** then browse through 18 factors to assess their value for further *** managerial implications are also discussed.

关键词： Big data two-sided market pricing information goods case study

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：