检索结果-内蒙古大学图书馆

International Conference on Signal processing and Communication Systems, ICSPCS

作者： Nicholas Jaton Sohan Gyawali Yi Qian Department of Electrical and Computer Engineering University of Nebraska-Lincoln Omaha NE USA Department of Technology Systems East Carolina University Greenville NC USA

As modern vehicular communication systems advance, the demand for robust security measures becomes increasingly critical. A misbehavior detection systems (MDS) is a tool developed to detect if a vehicular network is being attacked so that the system can take steps to mitigate harm from the attacker. Vehicular communication systems face significant risks from distributed denial of service (DDoS) attacks. During a DDoS attack, multiple nodes are used to flood the target with an overwhelming amount of communication packets. In this paper, we first survey the current MDS literature and how it is used to detect and mitigate DDoS attacks. We then propose a new distributed multilayer perceptron classifier (MLPC) for DDoS detection and evaluate the performance of the proposed detection scheme in vehicular communication systems. For the evaluations using simulations, two specific implementations of the attacks are conducted. Apache Spark is then used to create the distributed MLPC. The median F1-score for this MLPC method was 95%. The proposed method outperformed linear regression and support vector machines, which achieved 89% and 88% respectively, but is unable to perform better than random forests and gradient boosted trees which both achieved a 97% F1-score. Using Amazon Web Services (AWS), it is determined that model training and detection time are not significantly increased with the inclusion of additional nodes after three nodes including the master.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Early straggler tasks detection by recurrent neural network in a heterogeneous environment

引用

APPLIED INTELLIGENCE 2023年第7期53卷 7369-7389页

作者： Bawankule, Kamalakant Laxman Dewang, Rupesh Kumar Singh, Anil Kumar Motilal Nehru Natl Inst Technol Allahabad Dept Comp Sci & Engn Pryagraj Uttar Pradesh India

Heterogeneity is common in parallel and distributed environments used for extensive computations such as MapReduce. Stragglers are the tasks that are running on inferior performing nodes in the cluster. Early detection of stragglers is always challenging in such environments. In the previously proposed approaches, late detection of straggler tasks and estimation of time to end (TTE) for all the tasks running in a heterogeneous environment delays the entire job execution. Early straggler detection help to speculate a task at the early stages of task execution which indirectly improves the complete job execution. This article proposed early straggler detection by a recurrent neural network (ESDRNN) that collects the task and node information every three seconds from ApplicationMaster to train the RNN. It classifies the straggler tasks pretty early by RNN, between thirty to forty seconds of task execution, and transfers a list of classified tasks to an agent running on ResourceManager. RNN is a type of artificial neural network that is prevalent for processing sequential time-series data. Then, the agent predicts the TTE of these classified tasks by the Autoregressive integrated moving average (ARIMA) model. Finally, it sorts and refreshes the list with higher TTE after every ten seconds and speculates the tasks for the early completion of the MapReduce job. This proposed technique's performance is evaluated on the HiBench benchmark suite of Hadoop's most popular benchmark. Finally, compared with the default speculation technique and different techniques, the proposed speculation technique detects the stragglers early within 35 to 40 seconds of task execution. As a result, it decreases the job execution time by an average of 21% to 38% significantly for different workloads in a heterogeneous Hadoop cluster.

关键词： ANN (Artificial neural network) ESDRNN (Early straggler detection by a recurrent neural network) Hadoop Heterogeneous environment MapReduce RNN (Recurrent neural network) TTE (Time to end)

来源：评论

学校读者我要写书评

暂无评论

Explainable generalized additive neural networks with independent neural network training (vol 34, 6, 2024)

引用

STATISTICS AND COMPUTING 2024年第5期34卷 1-11页

作者： Ortega-Fernandez, Ines Sestelo, Marta Villanueva, Nora M. Galician Res & Dev Ctr Adv Telecommun GRADIANT Carretera Vilar 56-58 Vigo 36214 Spain CITMAga Santiago De Compostela 15782 Spain Univ Vigo Dept Stat & OR SiDOR Res Grp Vigo 36310 Spain Univ Vigo Escola Enxenaria Telecomunicac Vigo 36310 Spain

neural networks are one of the most popular methods nowadays given their high performance on diverse tasks, such as computer vision, anomaly detection, computer-aided disease detection and diagnosis or natural language processing. While neural networks are known for their high performance, they often suffer from the so-called “black-box” problem, which means that it is difficult to understand how the model makes decisions. We introduce a neural network topology based on Generalized Additive Models. By training an independent neural network to estimate the contribution of each feature to the output variable, we obtain a highly accurate and explainable deep learning model, providing a flexible framework for training Generalized Additive neural networks which does not impose any restriction on the neural network architecture. The proposed algorithm is evaluated through different simulation studies with synthetic datasets, as well as a real-world use case of distributed Denial of Service cyberattack detection on an Industrial Control System. The results show that our proposal outperforms other GAM-based neural network implementations while providing higher interpretability, making it a promising approach for high-risk AI applications where transparency and accountability are crucial.

关键词： Interpretable deep learning Generalized additive models Interpretable AI neural network GAM Explainable AI

来源：评论

学校读者我要写书评

暂无评论

Object recognition using cognition based decision tree clustering in multi-level artificial neural network classifier and self similarity as feature criterion

引用

MULTIMEDIA TOOLS AND APPLICATIONS 2024年第34期83卷 81591-81619页

作者： Kumar, Upendra Inst Engn & Technol Lucknow India Dr A P J Abdul Kalam Tech Univ Lucknow Uttar Pradesh India

This work showed the capability of handling large number of classes for classification with human cognition inspired methods. A cognition based techniques for both feature extraction, (self-similarity feature, Intensity Level Multi Fractal Dimension (ILMFD)) as well as classification purpose (decision tree clustering based multi-level Artificial neural network classifier-MLANN-DTC) were employed to implement facial recognition based object detection system. A DTC based approach reduces the search space time and also provides opportunity for very less amount of classes (a smaller part of the large number of classes) to be handled by the respective classifier for classification. It also mimics fast recognition capability of humans. In this work, two different databases were used for experiment, first one is our own collected facial images from rotation based video clips (117 persons and 40 facial images per person) named as NS database, and other is standard ORL database (40 persons and 10 facial images per person). In pre-processing step, the facial images were segmented to obtain facial part using context window based texture of pixels (CWTP) & back-propagation neural network (BPNN) based model and then a scale and rotation independent ILMFD feature was computed from each segmented image. Further, a combination of K-means and hierarchal clustering was used to build super classes. All classes' data were distributed among these 6 super classes (heuristically chosen) for own NS database and 3 for ORL database as per their similarity based on ILMFD features. Multi-level ANNs models were employed for all super classes and further their classification results were fed into decision clustering based model to obtain fine-tuned results, which showed significant improvement in terms of classification efficiency. This approach believes in center tendency of largest cluster to refer the actual class decision from multiple decisions obtain corresponding to multiple input data of

关键词： Human cognition Backpropogation neural network Decision Tree Clustering Decision Clustering Intensity Level Based Multifractal Dimension (ILMFD) Multi-level Artificial neural network (MLANN) Context Window Based Texture of Pixels (CWTP)

来源：评论

学校读者我要写书评

暂无评论

distributed Stochastic Gradient Tracking Algorithm With Variance Reduction for Non-Convex Optimization

引用

IEEE TRANSACTIONS ON neural networkS AND LEARNING SYSTEMS 2023年第9期34卷 5310-5321页

作者： Jiang, Xia Zeng, Xianlin Sun, Jian Chen, Jie Beijing Inst Technol Sch Automat Key Lab Intelligent Control & Decis Complex Syst Beijing 100081 Peoples R China Beijing Inst Technol Chongqing Innovat Ctr Chongqing 401120 Peoples R China Tongji Univ Sch Elect & Informat Engn Shanghai 200082 Peoples R China

This article proposes a distributed stochastic algorithm with variance reduction for general smooth non-convex finite-sum optimization, which has wide applications in signal processing and machine learning communities. In distributed setting, a large number of samples are allocated to multiple agents in the network. Each agent computes local stochastic gradient and communicates with its neighbors to seek for the global optimum. In this article, we develop a modified variance reduction technique to deal with the variance introduced by stochastic gradients. Combining gradient tracking and variance reduction techniques, this article proposes a distributed stochastic algorithm, gradient tracking algorithm with variance reduction (GT-VR), to solve large-scale non-convex finite-sum optimization over multiagent networks. A complete and rigorous proof shows that the GT-VR algorithm converges to the first-order stationary points with O(1/k) convergence rate. In addition, we provide the complexity analysis of the proposed algorithm. Compared with some existing first-order methods, the proposed algorithm has a lower O(PM epsilon b;(1)) gradient complexity under some mild condition. By comparing state-of-the-art algorithms and GT-VR in numerical simulations, we verify the efficiency of the proposed algorithm.

关键词： Signal processing algorithms Optimization Convergence Random variables distributed algorithms Machine learning distributed databases Complexity analysis distributed algorithm non-convex finite-sum optimization stochastic gradient variance reduction

来源：评论

学校读者我要写书评

暂无评论

An Experimental Comparison of Partitioning Strategies for distributed Graph neural network Training

arXiv

引用

arXiv 2023年

作者： Merkel, Nikolai Stoll, Daniel Mayer, Ruben Jacobsen, Hans-Arno Technical University of Munich Germany University of Bayreuth Germany University of Toronto Canada

Recently, graph neural networks (GNNs) have gained much attention as a growing area of deep learning capable of learning on graph-structured data. However, the computational and memory requirements for training GNNs on large-scale graphs make it necessary to distribute the training. A prerequisite for distributed GNN training is to partition the input graph into smaller parts that are distributed among multiple machines of a compute cluster. Although graph partitioning has been studied with regard to graph analytics and graph databases, its effect on GNN training performance is largely unexplored. As a consequence, it is unclear whether investing computational efforts into high-quality graph partitioning would pay off in GNN training scenarios. In this paper, we study the effectiveness of graph partitioning for distributed GNN training. Our study aims to understand how different factors such as GNN parameters, mini-batch size, graph type, features size, and scale-out factor influence the effectiveness of graph *** conduct experiments with two different GNN systems using vertex and edge partitioning. We found that high-quality graph partitioning is a very effective optimization to speed up GNN training and to reduce memory consumption. Furthermore, our results show that invested partitioning time can quickly be amortized by reduced GNN training time, making it a relevant optimization for most GNN scenarios. Compared to research on distributed graph processing, our study reveals that graph partitioning plays an even more significant role in distributed GNN training, which motivates further research on the graph partitioning problem. © 2023, CC BY-NC-ND.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Singular Value Manipulating: An Effective DRL-Based Adversarial Attack on Deep Convolutional neural network

引用

neural processing LETTERS 2023年第9期55卷 12459-12480页

作者： He, Shuai Fu, Cai Feng, Guanyun Lv, Jianqiang Deng, Fengyang Huazhong Univ Sci & Technol Hubei Engn Res Ctr Big Data Secur Sch Cyber Sci & Engn Hubei Key Lab Distributed Syst Secur Wuhan 430000 Hubei Peoples R China

In recent years, deep convolutional neural networks (DCNNs) have become increasingly prevalent in image processing applications. However, DCNNs are vulnerable to adversarial attacks, which are generated by adding imperceptible perturbations to the input that can cause the network to misclassify the image. In this study, we propose a black-box transferable adversarial attack method. The goal is to enhance the understanding of the vulnerability of these networks. Meanwhile, it could help develop more robust defenses against such attacks. This attack efficiently generates adversarial examples by manipulating the singular value matrix instead of directly perturbing pixels with complex noise. We utilize soft actor-critic to explore an optimal perturbation strategy. We perform extensive evaluations with VOC 2012, MS Coco 2017 datasets on object detection models, the MNIST dataset on image classification models, as well as the TT-100K dataset on a real-world case study to evaluate the proposed singular value manipulating attack (SVMA). Comparison results demonstrate that SVMA achieves a consistent query efficiency and attack ability on both one-stage detector Yolo and two-stage detector Faster R-CNN. Additionally, our case study demonstrates the adversarial examples of SVMA are effective in real-world scenarios. In the end, we propose a defense against such attacks.

关键词： Black-box adversarial attack Object detection Deep convolutional neural network Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Optimization of energy consumption of oil refinery reboiler and condenser using neural network

引用

neural Computing and Applications 2024年第32期36卷 20193-20209页

作者： Farahbod, Farshad Department of Chemical Engineering Firoozabad Branch Islamic Azad University Firoozabad Iran

The distillation tower is a crucial component of the refining process. Its energy efficiency has been a major area of research, especially following the oil crisis. This study focuses on optimizing energy consumption in the Shiraz refinery’s distillation unit. The unit is simulated using ASPEN-HYSYS software. Simulation results are validated against real data to ensure model accuracy. The operational data aligns well with model predictions. Following the creation of a data bank using HYSYS software, the tower’s operating conditions are optimized using neural networks and MATLAB software. In this study, a neural network model is developed for the distillation tower. This modeling approach is cost-effective, does not require complex theories, and does not rely on prior system knowledge. Additionally, real-time modeling is achievable through parallel distributed processing. The findings indicate that the optimal feed tray is 9 and the optimal feed temperature is 283.5°C. Furthermore, the optimized number of trays in the distillation tower is 47. Results show that in optimal conditions, cold and hot energy consumption are reduced by approximately 9.7% and 10.8%, respectively. Moreover, implementing optimal conditions results in a reduction of hot energy consumption in the reboiler by 60,000 MW and a reduction of cold energy consumption in the condenser by 30,000 MW. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： MATLAB

来源：评论

学校读者我要写书评

暂无评论

Exploration of English speech translation recognition based on the LSTM RNN algorithm

引用

neural COMPUTING & APPLICATIONS 2023年第36期35卷 24961-24970页

作者： Yuan, Qiwei Dai, Yu Li, Guangming Shaoyang Univ Coll Foreign Language Shaoyang 422000 Hunan Peoples R China Hunan Univ Med Coll Int Educ Huaihua 418000 Hunan Peoples R China Shaoyang Univ Dept Mech & Energy Engn Shaoyang 422000 Hunan Peoples R China

In today's information society, the demand for intelligence is increasing daily. English speech translation recognition technology based on the LSTM (long short-term memory) recurrent neural network (RNN) algorithm is an important manifestations of computer intelligence. In recent years, many scholars have conducted research on speech translation recognition technology, including template matching and statistical pattern recognition. Each of these methods has its drawbacks. This paper discusses English speech recognition techniques by utilizing the basic RNN principles. Moreover, its application and construction in practice, which can provide some useful reference for future researchers, are analysed. LSTM RNN is an intelligent system that is different from traditional pattern recognition methods. The greatest difference is that it simulates the information processing of the human brain and realizes the intelligent information processing in a distributed manner. It has a variety of automatic recognition and extraction functions, such as storage, association, and retrieval, especially for speech translation and recognition problems with high perception ability. This new neural network recognition system has a strong scientific nature and can store sound information in a decentralized manner, similar to the human brain. The LSTM RNN has been widely used in the speech recognition field due to its excellent performance in extraction and classification. The study found that the recognition accuracy of the original RNN was generally maintained between 48 and 54%, and the data loss rate was relatively high. The accuracy rate of speech recognition based on LSTM RNN was as high as 94%, and the information storage efficiency was high, which greatly avoided repetitive processes. The voice data processing speed can be completed in 4.5 s at the fastest, which plays an important role in terms of mass satisfaction and social development needs.

关键词： LSTM (long short-term memory) Recurrent neural network algorithm English voice translator Translation recognition system

来源：评论

学校读者我要写书评

暂无评论

A Federated Learning Approach for Graph Convolutional neural networks 13

A Federated Learning Approach for Graph Convolutional Neural...

引用

13rd IEEE Sensor Array and Multichannel Signal processing Workshop (SAM)

作者： Campbell, Andrew Liu, Hang Scaglione, Anna Wu, Tong Cornell Univ Cornell Tech Dept Elect & Comp Engn Ithaca NY 14853 USA

ISBN: (纸本)9798350344820;9798350344813

In this paper we cast the problem of training a graph neural network based on labeled graph data in a "federated learning" scenario where different agents have access to data from a subset of the network nodes. The learning problem is not decomposable, therefore it does not lend itself to a straightforward mapping onto a distributed multi-agent protocol. We propose a multi-agent federated learning scheme which leverages the local and sparse structure of graph filters to limit the information sharing while emulating the performance of centralized training. Even though we preserve data locality and agent communication is restricted to the neighborhood level, the proposed method still converges in simulation.

关键词： Graph convolutional neural network multi-agent federated learning

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：