检索结果-内蒙古大学图书馆

Optimization of Deep Learning Model for Plant Disease Detection Using Particle Swarm Optimizer

computers, Materials & Continua 2022年第5期71卷 4019-4031页

作者： Ahmed Elaraby Walid Hamdy Madallah Alruwaili Department of Computer Science Faculty of Computers and InformationSouth Valley UniversityQenaEgypt Department of Math. &Computer Science Faculty of SciencePort Said UniversityEgypt College of Computer and Information Sciences Jouf UniversitySakakaAljoufKingdom of Saudi Arabia

Plant diseases are a major impendence to food security,and due to a lack of key infrastructure in many regions of the world,quick identification is still *** losses owing to illnesses are a severe problem for both large farming structures and rural communities,motivating our *** of the large range of diseases,identifying and classifying diseases with human eyes is not only time-consuming and labor intensive,but also prone to being mistaken with a high error *** learning-enabled breakthroughs in computer vision have cleared the road for smartphone-assisted plant disease and *** proposed work describes a deep learning approach for detection plant ***,we proposed a deep learning model strategy for detecting plant disease and classification of plant leaf *** our research,we focused on detecting plant diseases in five crops divided into 25 different types of classes(wheat,cotton,grape,corn,and cucumbers).In this task,we used a public image database of healthy and diseased plant leaves acquired under realistic *** our work,a deep convolutional neural model AlexNet and Particle Swarm optimization was trained for this task we found that the metrics(accuracy,specificity,Sensitivity,precision,and Fscore)of the tested deep learning networks achieves an accuracy of 98.83%,specificity of 98.56%,Sensitivity of 98.78%,precision of 98.67%,and F-score of 98.47%,demonstrating the feasibility of this approach.

关键词： Deep neural networks plant diseases detection classification AlexNet PSO

来源：评论

学校读者我要写书评

暂无评论

The Impact of GPU on Containerized computer Vision Applications on Edge Nodes 8

The Impact of GPU on Containerized Computer Vision Applicati...

引用

8th IEEE International Conference on Edge Computing and Communications, EDGE 2024

作者： Alqaisi, Osamah I. Tosun, Ali Saman Korkmaz, Turgay University of Texas at San Antonio Department of Computer Science San AntonioTX78249 United States The University of North Carolina at Pembroke Dep. of Math. and Computer Science NC28372 United States

ISBN: (纸本)9798350368499

Edge computing revolutionizes computer vision, but faces challenges due to computational demands and complexities in deployment and platform integration. In addressing these hurdles, container technology emerges as a crucial solution, particularly in mitigating deployment and security issues. However, the impact of GPU acceleration on containerized computer vision applications remains understudied. This paper explores containerized computer vision applications across ARM edge devices, analyzing performance on different CPUs and GPUs. Insights cover processing time, CPU usage, memory utilization, and application performance, highlighting the importance of ARM processor selection. Integrating CPU and GPU enhances execution time but leads to significant memory usage increases. Workload escalation strains the GPU, resulting in slowed processing, and its performance becomes comparable to that of the CPU. Furthermore, the performance of computer vision applications varies with different containers. These findings underscore the delicate balance between CPU and GPU utilization and the pivotal role of algorithm selection in optimizing containerized computer vision applications on edge devices. © 2024 IEEE.

关键词： ARM processors

来源：评论

学校读者我要写书评

暂无评论

A Robust Cold-Start Approach Using Phylogram 10

A Robust Cold-Start Approach Using Phylogram

引用

10th International Conference on Optimization and Applications, ICOA 2024

作者： Da Silva, Breno Caetano Franco, Jorge Luiz Branco, Hermes Manoel Galvão Castelo Tokuda, Eric Delbem, Alexandre Claudio Botazzo Federal Institute of Maranhão Computer Science Department Teresina Brazil Inst. of Math. and Computer Science University of São Paulo São Carlos Brazil Federal University of Piaui Electrical Engineering Department Teresina Brazil

ISBN: (纸本)9798350387353

In Active Learning (AL) applications, it is crucial to develop a model that efficiently represents data using a limited number of labeled samples. In this context, an oracle is often employed to label samples that are deemed representative and informative. However, one of the primary limitations of AL lies in the need for an initial set of labeled samples, which may not be representative of the entire data space. To address this issue, cold-start strategies are implemented with the goal of selecting an appropriate initial set of samples for labeling, ensuring that the initial model is robust and effective. We propose a novel approach - PhyIL - that harness the advantages of phylogram analysis and community detection techniques to smartly perform the cold-start selection in AL. The method facilitates the choice of samples that best represent the diversity and complexity of the data, which is fundamental for the subsequent success of the active learning process. We evaluated the proposed approach using nine real-world databases and nine traditional ML models. The findings indicate that PhyIL offers an effective and viable solution for the initial sample selection, when compared to the random selection. © 2024 IEEE.

关键词： Active learning

来源：评论

学校读者我要写书评

暂无评论

IDS-INT:Intrusion detection system using transformer-based transfer learning for imbalanced network traffic

引用

Digital Communications and Networks 2024年第1期10卷 190-204页

作者： Farhan Ullah Shamsher Ullah Gautam Srivastava Jerry Chun-Wei Lin School of Software Northwestern Polytechnical UniversityXian710072ShaanxiChina Knowledge Unit of Systems and Technology University of Management and TechnologySialkot 51040Pakistan Department of Math and Computer Science Brandon UniversityBrandonMB R7A 6A9Canada Department of Computer Science and Math Lebanese American UniversityBeirut1102Lebanon Western Norway University of Applied Sciences BergenNorway Research Centre for Interneural Computing China Medical UniversityTaichungTaiwanChina

A network intrusion detection system is critical for cyber security against llegitimate *** terms of feature perspectives,network traffic may include a variety of elements such as attack reference,attack type,a subcategory of attack,host information,malicious scripts,*** terms of network perspectives,network traffic may contain an imbalanced number of harmful attacks when compared to normal *** is challenging to identify a specific attack due to complex features and data imbalance *** address these issues,this paper proposes an Intrusion Detection System using transformer-based transfer learning for Imbalanced Network Traffic(IDS-INT).IDS-INT uses transformer-based transfer learning to learn feature interactions in both network feature representation and imbalanced ***,detailed information about each type of attack is gathered from network interaction descriptions,which include network nodes,attack type,reference,host information,***,the transformer-based transfer learning approach is developed to learn detailed feature representation using their semantic ***,the Synthetic Minority Oversampling Technique(SMOTE)is implemented to balance abnormal traffic and detect minority ***,the Convolution Neural Network(CNN)model is designed to extract deep features from the balanced network ***,the hybrid approach of the CNN-Long Short-Term Memory(CNN-LSTM)model is developed to detect different types of attacks from the deep *** experiments are conducted to test the proposed approach using three standard datasets,i.e.,UNsWNB15,CIC-IDS2017,and *** explainable AI approach is implemented to interpret the proposed method and develop a trustable model.

关键词： Network intrusion detection Transfer learning Features extraction Imbalance data Explainable AI Cybersecurity

来源：评论

学校读者我要写书评

暂无评论

Enhancing Missing Data Imputation for Migrants Data: A Neutrosophic Set-Based Machine Learning Approach

引用

Neutrosophic Sets and Systems 2025年 81卷 479-502页

作者： Abdo, Doaa A. Salama, A.A. Abdelmegaly, Alaa A. Mahmoud, Hanan Khadari Mahdi Applied statistics and insurance department faculty of commerce Mansoura university Mansoura Egypt Dept. of Math and Computer Sci. Faculty of Science Port Said Univ. Egypt Higher Institute of Advanced Management Sciences and Computers Al-Buhayrah Egypt Depatrment of applied statistics at the Nile Higher Institute of commercial sciences and Computer Technology in Mansoura Egypt

This study tackles the problem of missing data in migrant datasets by introducing a new framework that combines machine learning techniques with neutrosophic sets. These sets, which can represent uncertainty and ambiguity, are well-suited for managing the complex nature of missing information in sensitive fields like migration research. We test the effectiveness of KNN, SVM, decision tree, random forest, and Ada Boost algorithms on a migrant dataset, comparing their results using different imputation methods (mean/mode, model-based imputer (simple tree), and random values). Our research showed that our proposed approach, which used neutrosophic sets, improved imputation accuracy and strengthened model reliability. Our results underscored the potential of neutrosophic set-based machine learning for addressing missing data issues across various fields. © 2025, University of New Mexico. All rights reserved.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

A Theory of Link Prediction via Relational Weisfeiler-Leman on Knowledge Graphs 37

A Theory of Link Prediction via Relational Weisfeiler-Leman ...

引用

37th Conference on Neural Information Processing Systems, NeurIPS 2023

作者： Huang, Xingyue Romero, Miguel Ceylan, Ismail Ilkan Barceló, Pablo Department of Computer Science University of Oxford Oxford United Kingdom Department of Computer Science Universidad Católica de Chile Chile CENIA Chile Chile Inst. for Math. and Comp. Eng. Universidad Católica de Chile Chile IMFD Chile Chile

ISBN: (纸本)9781713899921

Graph neural networks are prominent models for representation learning over graph-structured data. While the capabilities and limitations of these models are well-understood for simple graphs, our understanding remains incomplete in the context of knowledge graphs. Our goal is to provide a systematic understanding of the landscape of graph neural networks for knowledge graphs pertaining to the prominent task of link prediction. Our analysis entails a unifying perspective on seemingly unrelated models and unlocks a series of other models. The expressive power of various models is characterized via a corresponding relational Weisfeiler-Leman algorithm. This analysis is extended to provide a precise logical characterization of the class of functions captured by a class of graph neural networks. The theoretical findings presented in this paper explain the benefits of some widely employed practical design choices, which are validated empirically. © 2023 Neural information processing systems foundation. All rights reserved.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Generalized paths and cycles in semicomplete multipartite digraphs

arXiv

引用

arXiv 2024年

作者： Bang-Jensen, Jørgen Wang, Yun Yeo, Anders Dept. of Math. and Computer Science University of Southern Denmark Odense Denmark Dept. of Math. and Computer Science University of Southern Denmark Odense Denmark School of Mathematics Shandong University Jinan250100 China Department of Mathematics and Applied Mathematics University of Johannesburg Auckland Park2006 South Africa

A digraph is semicomplete if it has no pair of non-adjacent vertices. It is complete if every pair of distinct vertices induce a 2-cycle. It is well-known and easy to show that even the following version of the directed travelling salesman problem is NP-complete: Given a strongly connected complete digraph D = (V, A), a cost function w: A → {0, 1} and a natural number K;decide whether D has a directed Hamiltonian cycle of cost at most K. We study the following variant of this problem for {0, 1}-weighted semicomplete digraphs where the set of arcs which have cost 1 form a collection of vertex-disjoint complete digraphs. A digraph is semicomplete multipartite if it can be obtained from a semicomplete digraph D by choosing a collection of vertex-disjoint subsets X1, . . ., Xc of V (D) and then deleting all arcs both of whose end-vertices lie inside some Xi. Let D be a semicomplete digraph with a cost function w as above, where w(a) = 1 precisely when a is an arc inside one of the subsets X1, . . ., Xc and let D∗ be the corresponding semicomplete multipartite digraph that we obtain by deleting all arcs inside the Xi’s. Then every cycle C of D corresponds to a generalized cycle Cg of D∗ which is either the cycle C itself if w(C) = 0 or a collection of two or more paths that we obtain by deleting all arcs of cost 1 on C. Similarly we can define a generalized path P g in a semicomplete multipartite digraph. The purpose of this paper is to study structural and algorithmic properties of generalized paths and cycles in semicomplete multipartite digraphs. This allows us to identify classes of directed {0, 1}-weighted TSP instances that can be solved in polynomial time as well as others for which we can get very close to the optimum in polynomial time. Along with these results we also show that two natural questions about properties of cycles meeting all partite sets in semicomplete multipartite digraphs are *** Codes 05C20, 05c38, 05c45, 05c85 © 2024, CC BY.

关键词： Cost functions

来源：评论

学校读者我要写书评

暂无评论

Adjacency Sketches in Adversarial Environments

arXiv

引用

arXiv 2023年

作者： Naor, Moni Pekel, Eugene Department of Computer Science and Applied Math Weizmann Institute of Science

An adjacency sketching or implicit labeling scheme for a family F of graphs is a method that defines for any n vertex G ∈ F an assignment of labels to each vertex in G, so that the labels of two vertices tell you whether or not they are adjacent. The goal is to come up with labeling schemes that use as few bits as possible to represent the labels. By using randomness when assigning labels, it is sometimes possible to produce adjacency sketches with much smaller label sizes, but this comes at the cost of introducing some probability of error. Both deterministic and randomized labeling schemes have been extensively studied, as they have applications for distributed data structures and deeper connections to universal graphs and communication complexity. The main question of interest is which graph families have schemes using short labels, usually O(log n) in the deterministic case or constant for randomized sketches. In this work we consider the resilience of probabilistic adjacency sketches against an adversary making adaptive queries to the labels. This differs from the previously analyzed probabilistic setting which is "one shot". We show that in the adaptive adversarial case the size of the labels is tightly related to the maximal degree of the graphs in F . This results in a stronger characterization compared to what is known in the non-adversarial setting. In more detail, we construct sketches that fail with probability Ε for graphs with maximal degree d using 2d log(1/Ε) bit labels and show that this is roughly the best that can be done for any specific graph of maximal degree d, e.g. a d-ary *** Codes 68W20, 68W40, 68Q87 Copyright © 2023, The Authors. All rights reserved.

关键词： Graphic methods

来源：评论

学校读者我要写书评

暂无评论

Grassroots Systems: Concept, Examples, Implementation and Applications

arXiv

引用

arXiv 2023年

作者： Shapiro, Ehud Department of Computer Science and Applied Math Weizmann Institute of Science Israel

Informally, a grassroots system is a distributed system that can have multiple instances, independent of each other and of any global resources, that can interoperate once interconnected. More formally, in a grassroots system the set of all correct behaviors of a set of agents P is strictly included in the set of the correct behaviors of P when embedded within a larger set of agents P′ ⊃ P: Included, meaning that members of P are correct to ignore outside agents indefinitely;strictly included, meaning that the interaction of members of P with outside agents is also a possible and correct behavior. Client-server/cloud computing systems are not grassroots, and neither are systems designed to have a single global instance (Bitcoin/Ethereum with hardwired seed miners/bootnodes), and systems that rely on a single global data structure (IPFS, DHTs). An example grassroots system would be a serverless smartphone-based social network supporting multiple independently-budding communities that can merge when a member of one community becomes also a member of another. Grassroots applications are potentially important as they may allow people to conduct their social, economic, civic, and political lives in the digital realm solely using the networked computing devices they own and operate (e.g., smartphones), free of third-party control, surveillance, manipulation, coercion, or rent seeking (e.g., by global digital platforms such as Facebook or Bitcoin). Here, we formalize the notion of grassroots systems and grassroots implementations;specify an abstract grassroots dissemination protocol;describe and prove an implementation of grassroots dissemination for the model of asynchrony;extend the implementation to mobile (address-changing) devices that communicate via an unreliable network (e.g. smartphones using UDP);and discuss how grassroots dissemination can realize applications that support digital sovereignty – grassroots social networking and grassroots currencies. The math.mat

关键词： Smartphones

来源：评论

学校读者我要写书评

暂无评论

On biased compression for distributed learning

The Journal of Machine Learning Research

引用

The Journal of Machine Learning Research 2023年第1期24卷 12974-13023页

作者： Aleksandr Beznosikov Samuel Horváth Peter Richtárik Mher Safaryan Computer Electrical and Math. Sciences and Engineering Division King Abdullah University of Science and Technology Thuwal KSA and Skolkovo Institute of Science and Technology Moscow Russia and School of Applied Mathematics and Informatics Moscow Institute of Physics and Technology Moscow Russia Computer Electrical and Math. Sciences and Engineering Division King Abdullah University of Science and Technology Thuwal KSA

In the last few years, various communication compression techniques have emerged as an indispensable tool helping to alleviate the communication bottleneck in distributed learning. However, despite the fact biased compressors often show superior performance in practice when compared to the much more studied and understood unbiased compressors, very little is known about them. In this work we study three classes of biased compression operators, two of which are new, and their performance when applied to (stochastic) gradient descent and distributed (stochastic) gradient descent. We show for the first time that biased compressors can lead to linear convergence rates both in the single node and distributed settings. We prove that distributed compressed SGD method, employed with error feedback mechanism, enjoys the ergodic rate $O\left( \delta L \exp[-\frac{\mu K}{\delta L}] + \frac{(C + \delta D)}{K\mu}\right)$, where δ ≥1 is a compression parameter which grows when more compression is applied, L and µ are the smoothness and strong convexity constants, C captures stochastic gradient noise (C = 0 if full gradients are computed on each node) and D captures the variance of the gradients at the optimum (D = 0 for over-parameterized models). Further, via a theoretical study of several synthetic and empirical distributions of communicated gradients, we shed light on why and by how much biased compressors outperform their unbiased variants. Finally, we propose several new biased compressors with promising theoretical guarantees and practical performance.

关键词： compression operators biased compressors distributed learning linear convergence error feedback

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：