检索结果-内蒙古大学图书馆

Continual Learning with a Memory of Non-similar Samples 1

17th International Conference on Bio-Inspired Computing: Theories and Applications, BIC-TA 2022

作者： Min, Qilang He, Juanjuan Yang, Liuyan Fu, Yue College of Computer Science and Technology Wuhan University of Science and Technology Wuhan430081 China Hubei Province Key Laboratory of Intelligent Information Processing and Real-Time Industrial System Wuhan China

ISBN: (数字)9789819915491

ISBN: (纸本)9789819915484

The replay method is an effective strategy to address the problem of catastrophic forgetting, a key challenge for continuous learning. However, the sample sets obtained by replay-based methods generally suffer from local data information deficiencies. This problem leads to an imbalance in the plasticity-stability of the model on older tasks. This paper proposes a novel method, called non-similar sample storage (NSS). Non-similar refers to the Euclidean distance for feature vectors of different samples being far. NSS extracts the feature vectors of the samples and then calculates the similarity of the feature vectors for each sample after the current task training. Samples that contribute less to the model classification effect among similar samples are iteratively deleted, and the subset of low-similarity samples is retained. Moreover, NSS reserves 30% of the storage space for saving samples near the center of the sample set. Low-similarity samples stored by NSS get larger losses during the replay process, leading to lower training effectiveness of the current task. This paper introduces a knowledge distillation strategy to solve this problem. A variable parameter was used to balance the classification loss of the new task with the distillation loss of the old task (NSS-D). Experimental results in CIFAR10 and imbalanced CIFAR10 show that NSS maximizes the data’s global information and can retain the model’s ability to recognize old tasks better. Compared with classical algorithms, NSS-D performs better on CIFAR100 (48.8%) and ImageNet-200 (36.7%). © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

GAN-Based Unsupervised Learning Approach to Generate and Detect Fake News

GAN-Based Unsupervised Learning Approach to Generate and Det...

引用

International Conference on Cyber Security, Privacy and Networking, ICSPN 2022

作者： Bhardwaj, Pranjal Yadav, Krishna Alsharif, Hind Aboalela, Rania Anwar VIT Vellore Vellore India National Institute of Technology Kurukshetra Kurukshetra India Faculty of Computer and Information Technology Umm AL-Qura University Makkah Saudi Arabia Department of Information System King Abdulaziz University Rabigh Saudi Arabia

ISBN: (纸本)9783031220173

Social media has grown into an increasingly popular means of disseminating information. Its massive growth has given evolution to fake news in misinformation and rumors, spreading very quickly. These days the generation of fake news is not only limited to the traditional method but is also extended to deep learning-based methods. The characteristics of fake news generated from these algorithms are very much identical to original news, which makes existing supervised machine learning algorithms difficult to detect these machine-generated fake news. Motivated by the problem, we have brought a fully unsupervised approach based on Autoencoder and GAN. With the help of an autoencoder, we have generated the high dimensional feature vector of news sentences which is later used by generators in GAN to create machine-generated fake news. The generated fake news is then identified with the real news with the help of a discriminator. We have tested our approach with the news dataset that contains about 30,000 news headlines. The obtained experimental results suggest that our approach is very reliable and can be very helpful in automating fake news detection. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Critical Factors of Immersive Learning in Higher Education: A systematic Literature Review 10

Critical Factors of Immersive Learning in Higher Education: ...

引用

10th International Conference on ICT for Smart Society, ICISS 2023

作者： Sunardi Meyliana Ramadhan, Arief Rosmansyah, Yusep Bina Nusantara University BINUS Graduate Program Doctor of Computer Science Computer Science Department Jakarta Indonesia School of Information System Bina Nusantara University Information System Departement Jakarta Indonesia School of Computing Telkom University Bandung Indonesia School of Electrical Engineering and Informatics Bandung Institute of Technology Bandung Indonesia

ISBN: (纸本)9798350339543

Immersive learning has gained significant attention with the rising trend of spatial computing, particularly in the after-pandemic era. Numerous research has explored the potential of immersive learning in higher education, primarily on the educational sector. However, prior research has frequently focused too narrowly on the effects of technology and neglected to address the crucial element influencing successful immersive learning in higher education. This study seeks to pinpoint the crucial element contributing to the development of immersive learning experiences. The methodology uses a systematic literature review (SLR) from 2018 up to 2023 to investigate the critical factors of immersive Learning in Higher Education. From the 728 papers initially retrieved, 274 were considered potential candidates, and ultimately, 86 articles were selected based on their relevance to the research question. The results reveal that the critical factors include learning design, technology, immersion, engagement, interactivity, and usability. Academic interests will benefit from this SLR's consequences as institutions create models for designing suitable immersive learning, especially within the context of higher education. © 2023 IEEE.

关键词： E-learning

来源：评论

学校读者我要写书评

暂无评论

Performance Analysis of IDS_MDL Algorithm to Predict Intrusion Detection for IoT Applications 40

Performance Analysis of IDS_MDL Algorithm to Predict Intrusi...

引用

40th National Radio Science Conference, NRSC 2023

作者： Mahmoud, Walaa Adel Fathi, Mohamad El-Badawy, Hesham Sadek, Rowayda Coll. of Comp. Information Technology Arab Academy for Science Technology and Maritime Transport Information System Department Cairo Egypt National Telecommunication Institute Cairo Egypt Helwan University Faculty of Computer and Artificial Intelligence Cairo Egypt

ISBN: (纸本)9798350311440

In the last year, many modern attacks have targeted Internet of Things (IoT) devices. IoT botnet attacks are the most common. Despite the plethora of detection and prevention systems for cyberattacks, a security mechanism for IoT settings is still required since these devices have certain restrictions that make it difficult to properly implement the detection system. These memory resources are limited to processing, storing, and calculating. An effective defense against cyberattacks is an intrusion detection system (IDS). To develop the rules for IDS_MDL-IoT, we used the contemporary botnet attack dataset. Because IoT devices have limited resources, it is difficult to employ all attack patterns in the initial anomaly dataset in traditional anomaly-based IDS. Therefore, in this study, the tested ML model resulting in NB has proven to be the best classification algorithm to classify the risk of attack with 99.7% accuracy predicting from a time-speed of 3 seconds, and the tested DL model resulting in RNN has proven to be the best classification algorithm to classify the risk of attack with 99% accuracy predicting from a time-speed of 8 seconds. © 2023 IEEE.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Optimizing Steering Angle Predictive Convolutional Neural Network for Autonomous Car

引用

computers, Materials & Continua 2022年第5期71卷 2285-2302页

作者： Hajira Saleem Faisal Riaz Asadullah Shaikh Khairan Rajab Adel Rajab Muhammad Akram Mana Saleh Al Reshan Control Automotive and Robotics Lab Affiliated lab of National Center of Robotics and Automation(NCRA HEC Pakistan)and with the Department of Computer Science and Information TechnologyMirpur University of Science and Technology(MUST)Mirpur Azad Kashmir10250Pakistan College of Computer Science and Information System Najran UniversityNajran61441Saudi Arabia College of Computer Science and Engineering University of South FloridaTampa33620United States

Deep learning techniques,particularly convolutional neural networks(CNNs),have exhibited remarkable performance in solving visionrelated problems,especially in unpredictable,dynamic,and challenging *** autonomous vehicles,imitation-learning-based steering angle prediction is viable due to the visual imagery comprehension of *** this regard,globally,researchers are currently focusing on the architectural design and optimization of the hyperparameters of CNNs to achieve the best *** has proven the superiority of metaheuristic algorithms over the manual-tuning of ***,to the best of our knowledge,these techniques are yet to be applied to address the problem of imitationlearning-based steering angle ***,in this study,we examine the application of the bat algorithm and particle swarm optimization algorithm for the optimization of the CNN model and its hyperparameters,which are employed to solve the steering angle prediction *** validate the performance of each hyperparameters’set and architectural parameters’set,we utilized the Udacity steering angle dataset and obtained the best results at the following hyperparameter set:optimizer,Adagrad;learning rate,0.0052;and nonlinear activation function,exponential linear *** per our findings,we determined that the deep learning models show better results but require more training epochs and time as compared to shallower *** show the superiority of our approach in optimizing CNNs through metaheuristic algorithms as compared with the manual-tuning *** testing was also performed using the model trained with the optimal architecture,which we developed using our approach.

关键词： Bat algorithm convolutional neural network hyperparameters metaheuristic optimization algorithm steering angle prediction

来源：评论

学校读者我要写书评

暂无评论

Contrastive Learning for Robust Android Malware Familial Classification

引用

IEEE Transactions on Dependable and Secure Computing 2022年 1-14页

作者： Wu, Yueming Dou, Shihan Zou, Deqing Yang, Wei Qiang, Weizhong Jin, Hai National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Hubei Engineering Research Center on Big Data Security School of Cyber Science and Engineering Huazhong University of Science and Technology Wuhan China Shanghai Key Laboratory of Intelligent Information Processing School of Computer Science Fudan University Shanghai China University of Texas at Dallas Dallas USA National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China

Due to its open-source nature, Android operating system has been the main target of attackers to exploit. Malware creators always perform different code obfuscations on their apps to hide malicious activities. Features extracted from these obfuscated samples through program analysis contain many useless and disguised features, which leads to many false negatives. To address the issue, in this paper, we demonstrate that obfuscation-resilient malware family analysis can be achieved through contrastive learning. The key insight behind our analysis is that contrastive learning can be used to reduce the difference introduced by obfuscation while amplifying the difference between malware and other types of malware. Based on the proposed analysis, we design a system that can achieve robust and interpretable classification of Android malware. To achieve robust classification, we perform contrastive learning on malware samples to learn an encoder that can automatically extract robust features from malware samples. To achieve interpretable classification, we transform the function call graph of a sample into an image by centrality analysis. Then the corresponding heatmaps can be obtained by visualization techniques. These heatmaps can help users understand why the malware is classified as this family. We implement IFDroid and perform extensive evaluations on two datasets. Experimental results show that IFDroid is superior to state-of-the-art Android malware familial classification systems. Moreover, IFDroid is capable of maintaining a 98.4% F1 on classifying 69,421 obfuscated malware samples. IEEE

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Deep Learning-Based Real-Time Face Mask Detection for Human Using Novel YOLOv2 with Higher Accuracy 5th

Deep Learning-Based Real-Time Face Mask Detection for Human ...

引用

5th International Conference on computer and Communication Technologies, IC3T 2023

作者： Tun, Nay Kyi Myat, Aye Min Oo, Htet Ne Thu, Amar Myo Win, Kyaw Thu Faculty of Computer System and Technology Myanmar Information Institute of Technology Mandalay Myanmar Pyin Oo Lwin Myanmar Chitkara University Institute of Engineering and Technology Chitkara University Punjab Rajpura India Electronic Department Defence Services Academy Pyin Oo Lwin Myanmar

ISBN: (纸本)9789819997060

A novel deep learning-based face mask detection system is designed to enhance public safety in various environments. With the ongoing global health concerns, the need for efficient and accurate methods to identify individuals wearing or not wearing face masks has become crucial. By utilizing convolutional neural networks (CNNs) and transfer learning techniques, proposed model achieves impressive accuracy while maintaining high-speed processing capabilities. This paper outlines the architecture, training process, and performance evaluation of the proposed deep learning-based face mask detection system, highlighting its promising role in contributing to a safer and healthier society. The development of a vision-based safety system, the transfer of a small YOLO object detection model, and the creation of a CNN-based classification model are the key objectives of this study. According to experimental findings, proposed system is capable of real-time face mask detection and classification with an accuracy of over 98%. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

BHJO: A Novel Hybrid Metaheuristic Algorithm Combining the Beluga Whale, Honey Badger, and Jellyfish Search Optimizers for Solving Engineering Design Problems

引用

computer Modeling in Engineering & Sciences 2024年第10期141卷 219-265页

作者： Farouq Zitouni Saad Harous Abdulaziz S.Almazyad Ali Wagdy Mohamed Guojiang Xiong Fatima Zohra Khechiba Khadidja Â Kherchouche Department of Computer Science and Information Technology Kasdi Merbah UniversityP.O.Box 30000OuarglaAlgeria Department of Computer Science College of Computing and InformaticsUniversity of SharjahP.O.Box 27272SharjahUnited Arab Emirates Department of Computer Engineering College of Computer and Information SciencesKing Saud UniversityP.O.Box 51178Riyadh11543Saudi Arabia Operations Research Department Faculty of Graduate Studies for Statistical ResearchCairo UniversityGiza12613Egypt Applied Science Research Center Applied Science Private UniversityAmman11937Jordan Guizhou Key Laboratory of Intelligent Technology in Power System College of Electrical EngineeringGuizhou UniversityGuiyang550025China

Hybridizing metaheuristic algorithms involves synergistically combining different optimization techniques to effectively address complex and challenging optimization *** approach aims to leverage the strengths of multiple algorithms,enhancing solution quality,convergence speed,and robustness,thereby offering a more versatile and efficient means of solving intricate real-world optimization *** this paper,we introduce a hybrid algorithm that amalgamates three distinct metaheuristics:the Beluga Whale Optimization(BWO),the Honey Badger Algorithm(HBA),and the Jellyfish Search(JS)*** proposed hybrid algorithm will be referred to as *** this fusion,the BHJO algorithm aims to leverage the strengths of each *** this hybridization,we thoroughly examined the exploration and exploitation capabilities of the BWO,HBA,and JS metaheuristics,as well as their ability to strike a balance between exploration and *** meticulous analysis allowed us to identify the pros and cons of each algorithm,enabling us to combine them in a novel hybrid approach that capitalizes on their respective strengths for enhanced optimization *** addition,the BHJO algorithm incorporates Opposition-Based Learning(OBL)to harness the advantages offered by this technique,leveraging its diverse exploration,accelerated convergence,and improved solution quality to enhance the overall performance and effectiveness of the hybrid ***,the performance of the BHJO algorithm was evaluated across a range of both unconstrained and constrained optimization problems,providing a comprehensive assessment of its efficacy and applicability in diverse problem ***,the BHJO algorithm was subjected to a comparative analysis with several renowned algorithms,where mean and standard deviation values were utilized as evaluation *** rigorous comparison aimed to assess the performance of the BHJOalgorithmabout its counterparts,shedd

关键词： Global optimization hybridization of metaheuristics beluga whale optimization honey badger algorithm jellyfish search optimizer chaotic maps opposition-based learning

来源：评论

学校读者我要写书评

暂无评论

A Hierarchical Scheme for Video-Based Person Re-identification Using Lightweight PCANet and Handcrafted LOMO Features

引用

Chinese Journal of Electronics 2021年第2期30卷 289-295页

作者： LI Youjiao ZHUO Li LI Jiafeng ZHANG Jing Beijing Key Laboratory of Computational Intelligence and Intelligent System Beijing University of Technology Faculty of Information Technology College of Micro-electronics Beijing University of Technology College of Computer Science and Technology Shandong University of Technology

A two-level hierarchical scheme for video-based person re-identification(re-id) is presented, with the aim of learning a pedestrian appearance model through more complete walking cycle extraction. Specifically,given a video with consecutive frames, the objective of the first level is to detect the key frame with lightweight Convolutional neural network(CNN) of PCANet to reflect the summary of the video content. At the second level, on the basis of the detected key frame, the pedestrian walking cycle is extracted from the long video sequence. Moreover,local features of Local maximal occurrence(LOMO) of the walking cycle are extracted to represent the pedestrian's appearance information. In contrast to the existing walking-cycle-based person re-id approaches, the proposed scheme relaxes the limit on step number for a walking cycle, thus making it flexible and less affected by noisy frames. Experiments are conducted on two benchmark datasets: PRID 2011 and i LIDS-VID. The experimental results demonstrate that our proposed scheme outperforms the six state-of-art video-based re-id methods, and is more robust to the severe video noises and variations in pose,lighting, and camera viewpoint.

关键词： video surveillance image sequences two-level hierarchical scheme video content local features image representation video-based re-id methods pedestrian walking cycle video signal processing convolutional neural nets local maximal occurrence noisy frames lightweight PCANet complete walking cycle extraction pedestrian appearance information walking-cycle-based person re-id approaches LOMO features lightweight convolutional neural network learning (artificial intelligence) pedestrian appearance model consecutive frames video noises video-based person re-identification iLIDS-VID benchmark dataset LOMO PRID 2011 benchmark dataset video sequence image matching detected key frame object detection

来源：评论

学校读者我要写书评

暂无评论

DD-Net: Dynamic Network Architecture for Optimized Curve Segmentation and Reduce Computational Redundancy 27th

DD-Net: Dynamic Network Architecture for Optimized Curve Seg...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Cao, Yunxiang Chen, Li Wang, Yubo Feng, Zhida Tian, Jing School of Computer Science and Technology Wuhan University of Science and Technology Wuhan China Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System Wuhan University of Science and Technology Wuhan China Institute of Systems Science National University of Singapore Singapore Singapore

ISBN: (纸本)9783031783821

Curvilinear Structure segmentation has numerous applications in various fields, including providing a better understanding of defects such as cracks on roads or walls, thereby assessing the structural safety of buildings. Numerous curve segmentation methods based on Convolutional Neural Networks(CNN) have been developed in recent years. However, preserving feature integrity remains a challenge. To address this issue, this paper introduces a double dynamic network, called DD-Net, which consists of a dynamic structure for training and dynamic masking for inference. Firstly, to enhance the nonlinearity of the model, a paradigm to auto-adjust the CNN structure via a trainable module is formed for better fine-grained feature extraction. This dynamic structure (DS) paradigm enables a trade-off to extract or discard the feature data in the model training. Due to the data heterogeneity between the training and inference data, models may contain some useless nodes that are less effective during inference. The dynamic masking (DM) is proposed to omit the useless nodes based on the score difference between the train and inference feature statistics, thereby reducing redundant computations. To further improve the model’s performance on thin-curve segmentation and to preserve feature integrity, we introduce a non-curve suppression (NCS) module. This module focuses on background information while considering foreground prediction to address noisy conditions. The experimental results show that our DD-Net achieves promising results on three benchmark datasets and outperforms state-of-the-art curve segmentation models. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Convolutional neural networks

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：