检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Wang, Shang Gao, Yansong Fu, Anmin Zhang, Zhi Zhang, Yuqing Susilo, Willy Liu, Dongxi School of Computer Science and Engineering Nanjing University of Science and Technology China CSIRO’s Data61 Australia University of Western Australia Australia University of Chinese Academy of Sciences China School of Computing and Information Technology University of Wollongong Australia

As a critical threat to deep neural networks (DNNs), backdoor attacks can be categorized into two types, i.e., source-agnostic backdoor attacks (SABAs) and source-specific backdoor attacks (SSBAs). Compared to traditional SABAs, SSBAs are more advanced in that they have superior stealthier in bypassing mainstream countermeasures that are effective against SABAs. Nonetheless, existing SSBAs suffer from two major limitations. First, they can hardly achieve a good trade-off between ASR (attack success rate) and FPR (false positive rate). Besides, they can be effectively detected by the state-of-the-art (SOTA) countermeasures (e.g., SCAn [40]). To address the limitations above, we propose a new class of viable source-specific backdoor attacks coined as CASSOCK. Our key insight is that trigger designs when creating poisoned data and cover data in SSBAs play a crucial role in demonstrating a viable source-specific attack, which has not been considered by existing SSBAs. With this insight, we focus on trigger transparency and content when crafting triggers for poisoned dataset where a sample has an attacker-targeted label and cover dataset where a sample has a ground-truth label. Specifically, we implement CASSOCKTrans that designs a trigger with heterogeneous transparency to craft poisoned and cover datasets, presenting better attack performance than existing SSBAs. We also propose CASSOCKCont that extracts salient features of the attacker-targeted label to generate a trigger, entangling the trigger features with normal features of the label, which is stealthier in bypassing the SOTA defenses. While both CASSOCKTrans and CASSOCKCont are orthogonal, they are complementary to each other, generating a more powerful attack, called CASSOCKComp, with further improved attack performance and stealthiness. To demonstrate their viability, we perform a comprehensive evaluation of the three CASSOCK-based attacks on four popular datasets (i.e., MNIST, CIFAR10, GTSRB and LFW) and three

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Skin Cancer Classification of Dermoscopic Images using Deep Learning

Skin Cancer Classification of Dermoscopic Images using Deep ...

引用

Advancement in Electronics & Communication Engineering (AECE), International Conference on

作者： G Bindu Madhavi Sumithra M P Yamini Devi U S B K Mahalaxmi T Saravanan Pundru Chandra Shaker Reddy Department of CSE –AI &ML Geethanjali College of Engineering and Technology Hyderabad Telangana India Computer Science and Engineering School of Computing Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology Chennai Tamil Nadu India Computer Science and Engineering Vignana Bharathi Institute of Technology Hyderabad Telangana India Electronics Communication Engineering Aditya University Surampalem Andhra Pradesh India Department of Information Technology P. S. R. Engineering College Sivkasi Tamil Nadu India School of Computer Science and Artificial Intelligence SR University Warangal Telangana India

ISBN: (数字)9798350364729

ISBN: (纸本)9798350364736

Pigmented skin lesions must be identified in order to detect hazardous skin-related disorders, especially cancer. To improve the correctness of skin cancer recognition, image detection strategies and classification abilities can be utilized. Based on the HAM10000 dataset, which contains 10015 photos, this research paper makes use of the dataset. An augmentation has been carried out on a subset of the dataset in the proposed work. Compared to an approach with &without data augmentation, is more likely to learn unique traits and attributes. To make the model more accurate, data augmentation is often used. However, until the model is strong, it will not be able to produce meaningful results when tested with the testing data. This work's model is robust since it makes use of the k-fold cross-validation technique. Machine learning methods and models from Convolutional Neural Networks have been evaluated for their classification accuracy. After comparing the suggested work's use of several machine learning methods, we found that Convolutional Neural Network produced the best results in terms of accuracy. We achieved a maximum accuracy of 95.53% using the CNN in the suggested setup. Early detection of seven types of skin illness is made easier by the suggested work, which may be authenticated and treated properly by medicinal professionals.

关键词： Deep learning Accuracy Data augmentation Skin Data models Real-time systems Convolutional neural networks Lesions Skin cancer Testing

来源：评论

学校读者我要写书评

暂无评论

Evaluation of Wireless Sensor Networks Module using IoT Approach

Evaluation of Wireless Sensor Networks Module using IoT Appr...

引用

Artificial Intelligence and Smart Systems (ICAIS), International Conference on

作者： L. Anand Padmalal S J. Seetha R. Juliana PS Naveen Kumar Gayatri Parasa Department of Networking and Communications School of Computing SRM Institute of Science and Technology Kattanguluthur Department of Computer Science and Engineering Mangalam College of Engg Kerala Department of Computer Science and Business Systems Panimalar Engineering College Chennai Department of Information Technology Loyola-ICAM College of Engineering and Technology Chennai India Department of CSE-DATA SCIENCE St. Ann's College Of Engineering And Technology Chirala AP Department of Computer Science and Engineering Koneru Lakshmaiah Education Foundation Guntur Andhra Pradesh India

Microcomputers and medical devices with signal transceivers that operate on a specific radio display constitute the backbone of wireless sensor networks (WS Ns) that monitor environmental conditions (temperature, pressure, light, vibration levels, location). It is widely used in WAN sensor networks because of its flexible design and low setup fees. The u200b touch network allows for the connection of up to 65,000 devices, while the Intelligent sensors on other wireless networks are used to transfer data ports and assign wireless networks. Since the price of wireless solutions has been decreasing, and their functional capabilities have been growing, they are gradually replacing wired ones in telemetry data gathering systems and long- distance detecting communication. A deep learning model was used in this investigation to prevent the sensor nodes from manipulating data. Sensor nodes include a lot of parameters and estimations. If these projected data values are altered, network performance will suffer, and the node's lifetime will be reduced. Data security became a priority when the sensor nodes were distributed. This new method is 98.82% more efficient than the previous one.

关键词： Temperature sensors Wide area networks Vibrations Temperature measurement Wireless sensor networks Wireless networks Transceivers

来源：评论

学校读者我要写书评

暂无评论

Speed Master: Quick or Slow Play to Attack Speaker Recognition 39

Speed Master: Quick or Slow Play to Attack Speaker Recogniti...

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Ye, Zhe Zhang, Wenjie Ren, Ying Kang, Xiangui Yan, Diqun Ma, Bin Wang, Shiqi Guangdong Key Lab of Information Security School of Computer Science and Engineering Sun Yat-sen University Guangzhou China Faculty of Electrical Engineering and Computer Science Ningbo University Ningbo China Key Laboratory of Computing Power Network and Information Security Ministry of Education Qilu University of Technology Shandong Academy of Sciences Jinan China Department of Computer Science City University of Hong Kong Hong Kong

ISBN: (纸本)157735897X

Backdoor attacks pose a significant threat during the model’s training phase. Attackers craft pre-defined triggers to break deep neural networks, ensuring the model accurately classifies clean samples during inference yet erroneously classifies samples added with these triggers. Recent studies have shown that speaker recognition systems trained on large-scale data are susceptible to backdoor attacks. Existing attackers employ unnoticed ambient sounds as triggers. However, these sounds are not inherently part of the training samples themselves. In essence, triggers can be designed to maintain an intrinsic connection with the original speech to enhance stealthiness. Our paper presents a novel attack methodology named Speed Master, which undermines deep neural networks by manipulating the speed of speech samples. Specifically, we execute poison-only backdoor attacks using speed or tempo adjustment. Changes in speech rate have become a common occurrence, as seen on platforms that allow users to adjust playback speed. In real-world scenarios, people naturally adjust their speaking rate depending on the context. As a result, changes in a speaker’s speech rate are typically perceived as normal and are unlikely to raise suspicion. Furthermore, detecting such subtle adjustments becomes challenging for users without reference speech. Our comprehensive experiments demonstrate that Speed Master can achieve an ASR over 99% in the digital domain, with only a 0.6% poisoning rate. Additionally, we validate the feasibility of Speed Master in the real world and its resistance to typical defensive measures. Copyright © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Support Vector Machines Based Predictive Seizure Care using IoT-Wearable EEG Devices for Proactive Intervention in Epilepsy

Support Vector Machines Based Predictive Seizure Care using ...

引用

Computer, Communication and Control (IC4), International Conference on

作者： Porandla Srinivas M. Arulprakash M. Vadivel N Anusha G. Rajasekar C. Srinivasan Department of Information Technology Malla Reddy Institute of Engineering and Technology Hyderabad Telangana India Department of Computing Technologies SRM Institute of Science and Technology Kattankulathur Chenagalpattu Chennai Tamil Nadu India Department of Electronics and Communication Engineering Vidya Jyothi Institute of Technology Hyderabad Telangana India Department of IoT School School of Computer Science and Engineering Vellore Institute of Technology Vellore Tamil Nadu India Department of Computer Science and Engineering Vinayaka Mission’s Kirupananda Variyar Engineering College Vinayaka Mission’s Research Foundation (DU) Salem Tamil Nadu India Department of Computer Science and Engineering Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Saveetha University Chennai Tamil Nadu India

Epilepsy, a neurological illness that causes repeated seizures, can interfere with everyday life and needs prompt treatment. Internet of Things (IoT) wearable Electroencephalogram (EEG) devices and Support Vector Machines (SVM) for predictive analytics are used in this study to suggest a unique strategy for proactive seizure treatment. Wearable EEG devices feed real-time brain activity data. It uses SVM, a strong machine learning method, to build a prediction model using historical EEG data to identify and predict seizures. The prediction software analyses EEG data in real-time to detect pre-seizure patterns and initiate preventive treatments. The seizure prediction method uses SVM’s capacity to handle high-dimensional data and catch complicated patterns to improve accuracy and reliability. Healthcare practitioners and caregivers may get timely warnings and react efficiently thanks to the IoT infrastructure’s seamless connectivity between wearable devices and a centralized monitoring system. It discusses the ethical and privacy issues of installing such a system, stressing user permission and data protection. Pilot investigations show promising prediction accuracy and reaction time. SVM with IoT-wearable EEG sensors for predictive seizure care offers a forward-looking technique for enhancing epilepsy patients’ quality of life by enabling individualized and proactive treatments.

关键词：

来源：评论

学校读者我要写书评

暂无评论

HCF-Net: Hierarchical Context Fusion Network for Infrared Small Object Detection

arXiv

引用

arXiv 2024年

作者： Xu, Shibiao Zheng, ShuChen Xu, Wenhao Xu, Rongtao Wang, Changwei Zhang, Jiguang Teng, Xiaoqiang Li, Ao Guo, Li Artificial Intelligence Beijing University of Posts and Telecommunications China Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center Qilu University of Technology Shandong Academy of Sciences China State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Chinese Academy of Sciences China School of Artificial Intelligence University of Chinese Academy of Sciences China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science China School of Computer Science and Technology Harbin University of Science and Technology China

Infrared small object detection is an important computer vision task involving the recognition and localization of tiny objects in infrared images, which usually contain only a few pixels. However, it encounters difficulties due to the diminutive size of the objects and the generally complex backgrounds in infrared images. In this paper, we propose a deep learning method, HCF-Net, that significantly improves infrared small object detection performance through multiple practical modules. Specifically, it includes the parallelized patch-aware attention (PPA) module, dimension-aware selective integration (DASI) module, and multi-dilated channel refiner (MDCR) module. The PPA module uses a multi-branch feature extraction strategy to capture feature information at different scales and levels. The DASI module enables adaptive channel selection and fusion. The MDCR module captures spatial features of different receptive field ranges through multiple depth-separable convolutional layers. Extensive experimental results on the SIRST infrared single-frame image dataset show that the proposed HCF-Net performs well, surpassing other traditional and deep learning models. Code is available at https://***/zhengshuchen/HCFNet. Copyright © 2024, The Authors. All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

DL-HAR: Deep Learning-Based Human Activity Recognition Framework for Edge computing

引用

Computers, Materials & Continua 2020年第11期65卷 1033-1057页

作者： Abdu Gumaei Mabrook Al-Rakhami Hussain AlSalman Sk.Md.Mizanur Rahman Atif Alamri Research Chair of Pervasive and Mobile Computing King Saud UniversityRiyadh11543Saudi Arabia College of Computer and Information Sciences King Saud UniversityRiyadh11543Saudi Arabia Information and Communication Engineering Technology School of Engineering Technology and Applied ScienceCentennial CollegeTorontoCanada

Human activity recognition is commonly used in several Internet of Things applications to recognize different contexts and respond to *** learning has gained momentum for identifying activities through sensors,smartphones or even surveillance ***,it is often difficult to train deep learning models on constrained IoT *** focus of this paper is to propose an alternative model by constructing a Deep Learning-based Human Activity Recognition framework for edge computing,which we call *** goal of this framework is to exploit the capabilities of cloud computing to train a deep learning model and deploy it on less-powerful edge devices for *** idea is to conduct the training of the model in the Cloud and distribute it to the edge *** demonstrate how the DL-HAR can perform human activity recognition at the edge while improving efficiency and *** order to evaluate the proposed framework,we conducted a comprehensive set of experiments to validate the applicability of *** results on the benchmark dataset show a significant increase in performance compared with the state-of-the-art models.

关键词： Human activity recognition edge computing deep neural network recurrent neural network Docker

来源：评论

学校读者我要写书评

暂无评论

Time-aware Dynamic Graph Embedding for Asynchronous Structural Evolution

arXiv

引用

arXiv 2022年

作者： Yang, Yu Yin, Hongzhi Cao, Jiannong Chen, Tong Nguyen, Quoc Viet Hung Zhou, Xiaofang Chen, Lei The Department of Computing The Hong Kong Polytechnic University China The School of Information Technology and Electrical Engineering The University of Queensland Australia The Griffith University Australia The Department of Computer Science and Engineering The Hong Kong University of Science and Technology China

Dynamic graphs refer to graphs whose structure dynamically changes over time. Despite the benefits of learning vertex representations (i.e., embeddings) for dynamic graphs, existing works merely view a dynamic graph as a sequence of changes within the vertex connections, neglecting the crucial asynchronous nature of such dynamics where the evolution of each local structure starts at different times and lasts for various durations. To maintain asynchronous structural evolutions within the graph, we innovatively formulate dynamic graphs as temporal edge sequences associated with joining time of vertices (ToV) and timespan of edges (ToE). Then, a time-aware Transformer is proposed to embed vertices' dynamic connections and ToEs into the learned vertex representations. Meanwhile, we treat each edge sequence as a whole and embed its ToV of the first vertex to further encode the time-sensitive information. Extensive evaluations on several datasets show that our approach outperforms the state-of-the-art in a wide range of graph mining tasks. At the same time, it is very efficient and scalable for embedding large-scale dynamic graphs. © 2022, CC BY-NC-ND.

关键词： Graphic methods

来源：评论

学校读者我要写书评

暂无评论

Towards a Molecular Computer: Enabling Arithmetic Operations in Molecular Communication

arXiv

引用

arXiv 2025年

作者： Long, Jianqiao Zhang, Lei Wen, Miaowen Wang, Kezhi Krasnogor, Natalio Li, Jichun Interdisciplinary Computing and Complex BioSystems research group School of Computing Newcastle University Newcastle upon TyneNE4 5TG United Kingdom School of Electronic and Information Engineering South China University of Technology Guangzhou510641 China Department of Computer Science Brunel University of London Middlesex UxbridgeUB8 3PH United Kingdom

In current molecular communication (MC) systems, performing computational operations at the nanoscale remains challenging, restricting their applicability in complex scenarios such as adaptive biochemical control and advanced nanoscale sensing. To overcome this challenge, this paper proposes a novel framework that seamlessly integrates computation into the molecular communication process. The system enables arithmetic operations, namely addition, subtraction, multiplication, and division, by encoding numerical values into two types of molecules emitted by each transmitter to represent positive and negative values, respectively. Specifically, addition is achieved by transmitting non-reactive molecules, while subtraction employs reactive molecules that interact during propagation. The receiver demodulates molecular counts to directly compute the desired results. Theoretical analysis for an upper bound on the bit error rate (BER), and computational simulations confirm the system’s robustness in performing complex arithmetic tasks. Compared to conventional MC methods, the proposed approach not only enables fundamental computational operations at the nanoscale but also lays the groundwork for intelligent, autonomous molecular networks. Copyright © 2025, The Authors. All rights reserved.

关键词： Adaptive control systems

来源：评论

学校读者我要写书评

暂无评论

Sentimental Analysis of Movie Review Based on Naive Bayes and Random Forest Technique

Sentimental Analysis of Movie Review Based on Naive Bayes an...

引用

India Council International Subsections Conference (INDISCON), IEEE

作者： Vanshika Mittal Guru Prasad M S Harsh Kumar Vishwakarma Ganesh D R Chandrappa S Mangey Ram Dept. of Computer Science and Engg. Grapic Era (Deemed to be University) Dehradun India School of Computing and Information Technology REVA University Bengaluru India Dept. of Computer Science and Engg. JAIN (DEEMED-TO-BE UNIVERSITY) Bengaluru India

Rapid advancements in text classification algorithms have kept pace with the exponential rise of digital materials in recent years. In order to automatically extract expressive features, new machine learning algorithms have been suggested that take advantage of recent developments in deep learning techniques. Rapid progress in this area has spawned a multitude of ways for translating human speech into machine-readable data. Ad hoc pre-processing processes are utilized in conjunction with state-of-the-art language modelling algorithms; nevertheless, their presentation is generally glossed over in favor of a more thorough explanation of the classification stage. This work aims at building a model that is used to analyze sentiments, where accuracy is tested by taking data sets of positive and negative movie reviews. Proposed work is divided into 3 tasks: Data Extraction, Preprocessing and Modelling. We have used text classification techniques of Machine Learning by making changes in text vectorization using BOW (Bag Of Words), N-grams and TFIDF and modelling is built using Naïve Bayes and Random Forest algorithms. Our findings demonstrate that the suggested model provides superior performance in terms of accuracy, precision, recall, and f-measure. Using cutting-edge methods, proposed model achieved competitive results on the IMDB movie review dataset.

关键词：

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：