检索结果-内蒙古大学图书馆

Mitigating congestion in multi-agent traffic signal control: an efficient self-attention proximal policy optimization approach

引用

International Journal of Information Technology (Singapore) 2024年第4期16卷 2273-2282页

作者： Chergui, Oussama Sayad, Lamri Laboratory of Signals and Systems Analysis Faculty of Technology University of M′sila PO Box 166 Ichebilia M′sila 28000 Algeria Faculty of Mathematics and Computer Science University of M′sila PO Box 166 Ichebilia M′sila 28000 Algeria Laboratory of Informatics and its Applications of M′sila (LIAM) Faculty of Mathematics and Computer Science University of M′sila PO Box 166 Ichebilia M′sila 28000 Algeria

Traffic congestion is a persistent problem that effects cities worldwide, necessitating innovative solutions. This paper presents a novel traffic light control system using multi-agent proximal policy optimization with self-attention. Our approach outperforms traditional methods by 30% in reducing waiting times in high-traffic demand scenarios. By utilizing transfer learning and encoding mechanisms for dynamic input size adaptation, our approach enables scalability to larger networks without the need for costly training. This study underscores the potential of our approach as a dependable solution for addressing large-scale traffic congestion challenges. © 2023, The Author(s), under exclusive licence to Bharati Vidyapeeth's Institute of Computer applications and Management.

关键词： Deep reinforcement learning Multi-agent Proximal policy optimisation Self attention Traffic lights control Transfer learning

来源：评论

学校读者我要写书评

暂无评论

Lightweight and Secure Authentication Model for Vehicle to Everything (V2X) Communication Based on 5G Networks

引用

Journal of Mobile Multimedia 2022年第5期18卷 1399-1424页

作者： Houmer, Meriem Laqtib, Safaa Eddamiri, Siham Higher school of technologies Ibn Zohr University Dakhla Morocco Informatics and Applications Laboratory (IA) Department of Mathematics and Computer Science Faculty of Sciences Moulay Ismail University Meknes Morocco Department of Mathematics and Computer Science University Moulay Ismail ENSAM Meknes Morocco

Integrating the cellular network infrastructure into Intelligent Transport Systems (ITS) received attention from the research community and standards groups. This technology allows the invention and development of new applications, as well as the development of new vehicle mobility solutions and the enhancement of driver mobility in terms of safety, reliability, capacity, and quality. Cellular communications networks such as LTE-V2X and 5G NR, integrated into the Vehicle to Everything (V2X) architecture designed to support vehicular communications and deployed on a large scale, appear in particular to be a relevant solution. They could indeed guarantee reliable geographic distribution and acceptable performance (latency, bandwidth, packet loss). In this context, we propose a secure authentication scheme for 5G-based V2X communication. To achieve security requirements, our approach aims to provide a high degree of security in different vehicular communication (V2V, V2I, V2N) by using lightweight cryptographic algorithms in order to safely receive all keys and messages from RSU, vehicles, and the network. To validate our approach, we use the Automated Validation of Internet Security Protocols and applications (AVISPA) tool to achieve the security goals, also evaluate the performance according to the operational cost which demonstrates that our model has a less computational cost. © 2022 River Publishers.

关键词： 5G AVISPA ITS security V2X vehicular communications

来源：评论

学校读者我要写书评

暂无评论

CausAdv: A Causal-based Framework for Detecting Adversarial Examples

arXiv

引用

arXiv 2024年

作者： Debbi, Hichem Laboratory of Informatics and applications of M'sila Faculty of Mathematics and Computer Science University of M'sila PO Box 166 Ichebilia M'sila28000 Algeria

Deep learning has led to tremendous success in many real-world applications of computer vision, thanks to sophisticated architectures such as Convolutional neural networks (CNNs). However, CNNs have been shown to be vulnerable to crafted adversarial perturbations in *** inputs appear almost indistinguishable from natural images, yet they are incorrectly classified by CNN architectures. This vulnerability of adversarial examples has led researchers to focus on enhancing the robustness of deep learning models in general, and CNNs in particular, by creating defense and detection methods to distinguish adversarials inputs from natural ones. In this paper, we address the adversarial robustness of CNNs through causal reasoning. We propose CausAdv: a causal framework for detecting adversarial examples based on counterfactual reasoning. CausAdv learns causal and non-causal features of every input, and quantifies the counterfactual information (CI) of every filter of the last convolutional layer. Then we perform statistical analysis on the filters' CI of every sample, whether clan or adversarials, to demonstrate how adversarial examples indeed exhibit different CI distributions compared to clean samples. Our results show that causal reasoning enhances the process of adversarials detection without the need to train a separate detector. In addition, we illustrate the efficiency of causal explanations as a helpful detection technique through visualizing the causal features. The results can be reproduced using the code available in the repository: https://***/HichemDebbi/CausAdv. Copyright © 2024, The Authors. 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 Blind Digital Watermarking Approach Using Palmprint-Embedded Local Binary Patterns and Arnold Cat Map Transformations

A Blind Digital Watermarking Approach Using Palmprint-Embedd...

引用

International Conference on Pattern Analysis and Intelligent Systems (PAIS)

作者： Brahim Ferik Lakhdar Laimeche Abdallah Meraoumia Abdelkader Laouid Khaled Chait Muath AlShaikh Laboratory of Mathematics Informatics and Systems (LAMIS) University of Larbi Tebessi Tebessa Algeria Artificial Intelligence and its Applications Laboratory (LIAP) University of El Oued El Oued Algeria Department of Computer Science College of Computing and Informatics Saudi Electronic University Kingdom of Saudi Arabia

ISBN: (数字)9798350350265

ISBN: (纸本)9798350350272

The rapidly evolving healthcare domain and the rise of telemedicine and remote diagnostic techniques underscore the critical demand for secure, authentic, and reliable medical imaging solutions. This study introduces an innovative digital watermarking method designed to safeguard medical images against unauthorized alterations, employing biometric data—specifically, palmprints—as the watermark. Utilizing Local Binary Patterns (LBP) for feature extraction and combining Arnold Cat Map (ACM) with Discrete Cosine Transform (DCT) for image processing, this research implements a novel embedding strategy based on a 2 Least Significant Bits (2LSB) mechanism, facilitating a blind watermarking approach. The efficacy of the proposed method is validated through exceptional Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM) metrics, achieving values of 51.21675 dB and 0.9999, respectively. These metrics provide substantial evidence of the technique’s imperceptibility and robustness, employing state-of-the-art methodologies. The findings of this paper offer a compelling solution for enhancing the security of medical imagery within the telemedicine sector, promoting data integrity and supporting the verification and authentication processes.

关键词： Measurement PSNR Telemedicine Authentication Watermarking Feature extraction Robustness

来源：评论

学校读者我要写书评

暂无评论

Current knowledge about immunotherapy resistance for melanoma and potential predictive and prognostic biomarkers

引用

Cancer Drug Resistance 2024年第1期7卷 748-771页

作者： Lanni Song Yixin Yang Xuechen Tian Wenzhou Municipal Key Laboratory for Applied Biomedical and Bio-pharmaceutical Informatics Wenzhou-Kean UniversityWenzhou 325060ZhejiangChina Zhejiang Bioinformatics International Science and Technology Cooperation Center Wenzhou-Kean UniversityWenzhou 325060ZhejiangChina College of Science Mathematics and TechnologyWenzhou-Kean UniversityWenzhou 325060ZhejiangChina Dorothy and George Hennings College of Science Mathematics and TechnologyKean UniversityUnionNJ 07083USA

Melanoma still reaches thousands of new diagnoses per year,and its aggressiveness makes recovery challenging,especially for those with stage III/IV unresectable ***,emerging as a beacon of hope,stands at the forefront of treatments for advanced *** review delves into the various immunotherapeutic strategies,prominently featuring cytokine immunotherapy,adoptive cell therapy,immune checkpoint inhibitors,and *** these,immune checkpoint inhibitors,notably anti-programmed cell death-1(PD-1)and anti-cytotoxic T lymphocyte antigen-4(CTLA-4)antibodies,emerge as the leading ***,a significant subset of melanoma patients remains unresponsive to these inhibitors,underscoring the need for potent *** biomarkers have the potential to revolutionize the therapeutic landscape by facilitating the design of personalized treatments for patients with *** comprehensive review highlights the latest advancements in melanoma immunotherapy and potential biomarkers at the epicenter of recent research endeavors.

关键词： Melanoma immunotherapy resistance mechanism tumor microenvironment biomarkers

来源：评论

学校读者我要写书评

暂无评论

An analytical model for analyzing the value of information flow in the production chain model using regression algorithms and neural networks

Supply Chain Analytics

引用

Supply Chain Analytics 2023年 2卷

作者： Biyeme, Florent Mbakop, André Marie Chana, Anne Marie Voufo, Joseph National Advanced School of Engineering of Yaoundé (NASEY) Cameroon Laboratory of Civil and Mechanical Engineering Yaoundé Cameroon Department of Industrial and Mechanical Engineering Yaoundé Cameroon Laboratory of Mathematics Informatics Bio-informatics and Applications Yaoundé Cameroon Department of Computer Science Engineering Yaoundé Cameroon

Managing information flow has always been a challenging and critical driver of performance increase in manufacturing companies. Each bit of information related to the manufacturing process has an information flow value that can impact the process. Recent studies have focused on the traditional classification algorithms methods to analyze the value of information flow. In this research paper, we use regression algorithms to develop an analytics model for the value of information flow in manufacturing shop floors of developing countries. The analysis shows that the Artificial Neural Network (ANN) has the best regression coefficient score of 0.775 with a prediction error of 0.0125. The lowest regression coefficient score of 0.323 was for the Multi-Linear Regression (MLR) with a prediction error of 0.0556. These results help companies use regression algorithms effectively to analyze the value of information flows on the manufacturing chains. © 2023 The Authors

关键词： Artificial neural network Information flow management Production chain model Regression algorithms Value of information

来源：评论

学校读者我要写书评

暂无评论

Defending Against App-Layer DDoS: Advanced Machine Learning Security

Defending Against App-Layer DDoS: Advanced Machine Learning ...

引用

Innovative Research in Applied Science, Engineering and Technology (IRASET),Conference on

作者： Salim Salmi Lahcen Oughdir Anas El Affar Engineering Systems and Applications Laboratory Sidi Mohamed Ben Abdellah University ENSA Fez Morocco Department of Mathematics Physics and Informatics Sidi Mohamed Ben Abdellah University Taza FP Morocco

ISBN: (数字)9798350309508

ISBN: (纸本)9798350309515

Distributed denial-of-service attacks in the application layer (DDoS) pose a significant threat to online services, leading to significant financial losses and reputational damage. This intriguing research paper presents a novel automated detection approach to identify dangerous application layer DDoS attacks. Four machine learning algorithms, including random forest, decision tree, K-nearest neighbor and naive Bayes, were tested on three attack scenarios: benign (legitimate), DoS Slowloris and DoS Hulk using network traffic data feature extraction from deep packet inspection. The results showed that the Random Forest classifier excelled in all cases with incredible accuracy, precision recall, and F1 scores. The proposed approach provides a robust and efficient solution for mitigating application layer DDoS attacks, enhancing the security of online services, and reducing the impact of DDoS attacks.

关键词： Machine learning algorithms Focusing Telecommunication traffic Inspection Distributed denial-of-service attack Feature extraction Real-time systems

来源：评论

学校读者我要写书评

暂无评论

Using Crank-Nicolson Scheme for Continuous Hopfield Network Equilibrium 1

引用

2nd international conference Artificial Intelligence and Industrial applications, A2IA 2023

作者： Rbihou, Safae Joudar, Nour-Eddine En-Naimani, Zakariae Haddouch, Khalid Engineering Systems and Applications Laboratory ENSA Fez Sidi Mohamed Ben Abdellah University Fez Morocco Research Center STIS M2CS Department of Applied Mathematics and Informatics ENSAM Mohammed V University in Rabat Rabat Morocco Signaux Systèmes Distribués et Intelligence Artificielle Laboratory ENSET Mohammedia Hassan II University Casablanca Morocco

ISBN: (数字)9783031435201

ISBN: (纸本)9783031435195

The Continuous Hopfield Network (CHN) so far remained a concrete example of recurrent neural networks. The use of CHN in various applications is excellent, which generally leads to better results. The continuous Hopfield network has an energy function associated with it, and the network is bound to converge if the activity of each neuron with respect to time is given by a differential equation. The resolution of this differential equation of CHN in most works is based on the Euler method. In this work, we try to use another method for the solution of the CHN differential equation and rely on the Crank-Nicolson method. This work aims to improve and develop the performance of CHN using the Crank-Nicolson method. For this purpose, we have carried out an analytical and comparative study between the two methods of solving the CHN differential equation Euler and Crank-Nicolson applied to the problem of task assignment. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Differential equations

来源：评论

学校读者我要写书评

暂无评论

Simulation to predict the behaviour of a new seaport gate: An application in the Sines container terminal

引用

Procedia Computer Science 2025年 253卷 2409-2418页

作者： Raquel Gil Pereira Rui Borges Lopes Leonor Teixeira Center for Research & Development in Mathematics and Applications (CIDMA) Department of Economics Management Industrial Engineering and Tourism (DEGEIT) University of Aveiro Aveiro Portugal Institute of Electronics and Informatics Engineering of Aveiro (IEETA) Intelligent Systems Associate Laboratory (LASI) Department of Economics Management Industrial Engineering and Tourism (DEGEIT)University of Aveiro Aveiro Portugal

With rapid technological changes and globalized supply chains, intelligent and adaptable logistics systems are crucial, particularly at seaports. Container terminals face increasing volumes and demands, leading to potential truck congestion and long queues. Automation offers a solution by optimizing port performance and adapting to dynamic conditions, though port digitalization is complex due to diverse stakeholders and information flows. This work is part of the digital transformation of the Port of Sines - Portugal, specifically in the Smart Gate concept context. For this purpose, the Port of Sines Administration, together with other partners, has established the prerequisite of a pre-gate, which is currently being analysed from the point of view of its processes, the technologies to be included and the information models to be supported. This work aims to dynamically analyse the pre-gate model by testing different scenarios to propose improvements to the inherent processes using simulation. The study culminated in a discrete-event simulation model with Anylogic software to predict pre-gate behaviour and foresee potential issues. By testing various scenarios and evaluating their effects on key performance indicators, this model supports the continuous improvement of pre-gate processes.

关键词： port logistics container terminal smart gate truck congestion simulation Anylogic

来源：评论

学校读者我要写书评

暂无评论

A Review on recent research in information retrieval

引用

Procedia Computer Science 2022年 201卷 777-782页

作者： S. Ibrihich A. Oussous O. Ibrihich M. Esghir Mohammed V University in Rabat Faculty of Sciences Laboratory of Mathematics Computing and Applications Rabat Morocco Informatics Department Faculty of Sciences and Technologies of Mohammedia (FSTM) Hassan II University Casablanca Morocco

In this paper, we present a survey of modeling and simulation approaches to describe information retrieval basics. We investigate its methods, its challenges, its models, its components and its applications. Our contribution is twofold: on the one hand, reviewing the literature on discovery some search techniques that help to get pertinent results and reach an effective search, and on the other hand, discussing the different research perspectives for study and compare more techniques used in information retrieval. This paper will also shedding the light on some of the famous AI applications in the legal field.

关键词： Information Retrieval Intelligent Search IR models Data Mining Natural Language Processing

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：