检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Qiang, Wenwen Song, Zeen Gu, Ziyin Li, Jiangmeng Zheng, Changwen Sun, Fuchun Xiong, Hui The National Key Laboratory of Space Integrated Information System Institute of Software Chinese Academy of Sciences University of Chinese Academy of Sciences Beijing China The National Key Laboratory of Space Integrated Information System Department of Computer Science and Technology Tsinghua University Beijing China The Hong Kong University of Science and Technology China

Self-supervised learning (SSL) methods learn from unlabeled data and achieve high generalization performance on downstream tasks. However, they may also suffer from overfitting to their training data and lose the ability to adapt to new tasks. To investigate this phenomenon, we conduct experiments on various SSL methods and datasets and make two observations: (1) Overfitting occurs abruptly in later layers and epochs, while generalizing features are learned in early layers for all epochs;(2) Coding rate reduction can be used as an indicator to measure the degree of overfitting in SSL models. Based on these observations, we propose Undoing Memorization Mechanism (UMM), a plug-and-play method that mitigates overfitting of the pre-trained feature extractor by aligning the feature distributions of the early and the last layers to maximize the coding rate reduction of the last layer output. The learning process of UMM is a bi-level optimization process. We provide a causal analysis of UMM to explain how UMM can help the pre-trained feature extractor overcome overfitting and recover generalization. We also demonstrate that UMM significantly improves the generalization performance of SSL methods on various downstream tasks. The source code is to be released at https://***/ZeenSong/UMM. © 2024, CC BY.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Confident Information Coverage Reliability Evaluation for Sensor Networks of Openly Deployed ICP systems

IEEE Transactions on Industrial Cyber-Physical Systems

引用

IEEE Transactions on Industrial Cyber-Physical systems 2024年 2卷 565-574页

作者： Chen, Suning Yi, Yuanyuan Yi, Lingzhi Liu, Shenghao Deng, Xianjun Xia, Yunzhi Fan, Xiaoxuan Yang, Laurence T. Park, Jong Hyuk Huazhong University of Science and Technology Hubei Key Laboratory of Distributed System Security Hubei Engineering Research Center on Big Data Security The School of Cyber Science and Engineering Wuhan430074 China Huazhong University of Science and Technology School of Journalism and Information Communication Wuhan430074 China Zhongnan University of Economics and Law School of Information and Safety Engineering Wuhan430073 China Seoul National University of Technology Department of Computer Science and Engineering Seoul01811 Korea Republic of

Industrial Cyber-Physical (ICP) systems are integration of computation and physical processes to help achieve operational excellence. As sensors and actuators compose the openly deployed ICP systems and are often susceptible to interference, it puts forward higher requirements on reliability. Reliable coverage ensures sensors acquiring information and therefore coverage reliability of sensor network is studied from various perspectives. However, most existing reliability evaluation methods ignore the belief degree of coverage which influences the reliability to an extent. To evaluate the coverage reliability from the perspective of belief degree, we firstly define Confident Information Coverage Belief Reliability (CICBR) based on the Confident Information Coverage (CIC) model and D-S evidence theory, which comprehensively considers the coverage rate, interference from the environment, and belief degree of the result. Next, to evaluate the belief degree more reasonably, an Evidence-discount-based basic probability assignment generating method is proposed based on D-S theory to describe the interference in the belief degree. Moreover, a belief evaluation algorithm is proposed to calculate the belief degree of the coverage. Finally, an evidence-Discount-based Confident Information Coverage Belief Reliability Evaluation (D-CICBRE) algorithm is proposed to evaluate CICBR. The simulation result indicates that the proposed method can obtain higher and more reasonable reliability. Therefore, the proposed D-CICBRE algorithm can be used to evaluate the reliability of ICP systems more accurately and reasonably. © 2023 IEEE.

关键词： Sensor networks

来源：评论

学校读者我要写书评

暂无评论

Differences in Precipitation and Related Wind Dynamics and Moisture and Heat Features in Separate Areas of the South China Sea before and after Summer Monsoon Onset

引用

Advances in Atmospheric Sciences 2024年第8期41卷 1643-1660页

作者： Chunyan ZHANG Donghai WANG Kaifeng ZHANG Wanwen HE Yanping ZHENG Yan XU Guangdong Meteorological Data Center Guangdong Meteorological BureauGuangzhou 510640China School of Atmospheric Sciences Sun Yat-sen UniversityGuangdong Province Key Laboratory for Climate Change and Natural Disaster StudiesKey Laboratory of Tropical Atmosphere-Ocean SystemMinistry of EducationZhuhai 519082China Southern Marine Science and Engineering Guangdong Laboratory(Zhuhai) Zhuhai 519082China National Observation and Research Station of Coastal Ecological Environments in Macao Macao Environmental Research InstituteMacao University of Science and TechnologyMacao SAR 999078China Foshan Meteorological Service Foshan 528000China Ningjiang District Meteorological Bureau Songyuan 138000China

Using surface and balloon-sounding measurements, satellite retrievals, and ERA5 reanalysis during 2011–20, this study compares the precipitation and related wind dynamics, moisture and heat features in different areas of the South China Sea(SCS) before and after SCS summer monsoon onset(SCSSMO). The rainy sea around Dongsha(hereafter simply referred to as Dongsha) near the north coast, and the rainless sea around Xisha(hereafter simply referred to as Xisha) in the western SCS, are selected as two typical research subregions. It is found that Dongsha, rather than Xisha, has an earlier and greater increase in precipitation after SCSSMO under the combined effect of strong low-level southwesterly winds, coastal terrain blocking and lifting, and northern cold air. When the 950-h Pa southwesterly winds enhance and advance northward, accompanied by strengthened moisture flux, there is a strong convergence of wind and moisture in Dongsha due to a sudden deceleration and rear-end collision of wind by coastal terrain blocking. Moist and warm advection over Dongsha enhances early and deepens up to 200 h Pa in association with the strengthened upward motion after SCSSMO, thereby providing ample moisture and heat to form strong precipitation. However, when the 950-h Pa southwesterly winds weaken and retreat southward, Xisha is located in a wind-break area where strong convergence and upward motion centers move in. The vertical moistening and heating by advection in Xisha enhance later and appear far weaker compared to that in Dongsha, consistent with later and weaker precipitation.

关键词： Dongsha Xisha South China Sea summer monsoon onset precipitation wind dynamics moisture heat

来源：评论

学校读者我要写书评

暂无评论

Autonomous Traffic Prediction for LEO Satellite-Based IoT Based on Satellite Spatiotemporal Features Mapping

引用

IEEE Internet of Things Journal 2025年

作者： Gong, Lizeng Chen, Quan Yang, Lei Yin, Zhenglong Wang, Yi National University of Defense Technology State Key Laboratory of Space System Operation and Control College of Aerospace Science and Engineering Changsha410073 China Nanjing Research Institute of Electronics Technology Nanjing210013 China

The traffic prediction method for Low Earth Orbit (LEO) satellite-based Internet of Things (IoT) provides prior conditions for addressing challenges in resource allocation and management of LEO satellite-based IoT. However, traditional methods rely on the pre-generation of terrestrial centers, leading to substantial computational complexity and delays when applying prediction results to LEO satellite-based IoT. Therefore, achieving autonomous LEO satellite-based IoT traffic prediction is essential. Unlike the spatiotemporal features of terrestrial IoT, LEO satellite-based IoT traffic is influenced by both satellite dynamics and terrestrial user behavior. Fortunately, real IoT traffic information can be shared between adjacent satellites through Inter-satellite Links (ISL). Inspired by the above, this paper proposes, for the first time, an autonomous LEO satellite-based IoT traffic prediction method. A Spatiotemporal Fusion Neural Network (SFNN) employing a four-layer hybrid neural network architecture is designed to accommodate the unique spatiotemporal features of LEO satellite-based IoT traffic. The simulation results demonstrate that the proposed method has low computational complexity and achieves a better prediction performance than the five baseline methods in LEO constellations. Further simulations demonstrate that the proposed method achieves higher accuracy in large-scale LEO constellations, and the algorithm remains applicable under ISL instability and satellite failure conditions. © 2014 IEEE.

关键词： Resource allocation

来源：评论

学校读者我要写书评

暂无评论

Analysis on the causes of massive stranding of Yellow Sea green tide on Lianyungang and Rizhao coasts in 2022

引用

Journal of Oceanology and Limnology 2024年第3期42卷 816-830页

作者： Lingjuan WU Juan HUANG Yi DING Guiyan LIU Shuai HUANG Song GAO Chao YUAN Jiangling XU Ping WU Rui HUANG Ruobing WEN Jie MEI North China Sea Marine Forecasting and Hazard Mitigation Center Ministry of Natural ResourcesQingdao 266061China Shandong Key Laboratory of Marine Ecological Environment and Disaster Prevention and Mitigation Qingdao 266061China National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology Xi’an 710129China East China Sea Marine Forecasting and Hazard Mitigation Center Ministry of Natural ResourcesShanghai 200136China North China Sea Ecological Center Ministry of Natural ResourcesQingdao 266033China

In 2022,Yellow Sea green tide caused by Ulva prolifera reached a historic minimum and the coastal areas of Shandong Peninsula were less ***,the largest amount of seaweed biomass has been washed ashore on Lianyungang and Rizhao coasts since *** studied the development pattern of Yellow Sea green tide in 2022,and analyzed the key environmental factors on the growth and drifting,then discussed the possible reasons that resulted in the massive stranding of green tide biomass in Lianyungang and *** show under the combined influence of the east to southeast winds and currents with shoreward anomalies,green tide drifted to the coastal waters between Shandong and Jiangsu provinces and the distribution areas located westward compared with previous years(2008–2021).Floating *** rafts from the coastal waters of Binhai and Sheyang drifted continuously into the coastal waters of Lianyungang and Rizhao,providing important supplements for Yellow Sea green *** green tide in 2022 distributed close to the coastal waters,the abundant nutrients might support their continuous high growth *** addition,the amount of rainfall around Shandong Peninsula from late June to early July were significantly higher than in previous years,which might promote the development of green tide to some extent.

关键词： Yellow Sea green tide stranding biomass development pattern environmental factor

来源：评论

学校读者我要写书评

暂无评论

Towards Demystifying Android Adware: dataset and Payload Location 39

Towards Demystifying Android Adware: Dataset and Payload Loc...

引用

39th ACM/IEEE International Conference on Automated Software Engineering Workshops, ASEW 2024

作者： Wang, Chao Liu, Tianming Zhao, Yanjie Zhang, Lin Du, Xiaoning Li, Li Wang, Haoyu Monash University Clayton Australia Huazhong University of Science and Technology Wuhan China School of Cyber Science and Engineering Huazhong University of Science and Technology Hubei Key Laboratory of Distributed System Security Hubei Engineering Research Center on Big Data Security China The National Computer Emergency Response Beijing China Beihang University Beijing China

ISBN: (纸本)9798400712494

Adware represents a pervasive threat in the mobile ecosystem, yet its inherent characteristics have been largely overlooked by previous research. This work takes a crucial step towards demystifying Android adware. We present a comprehensive, well-annotated Android adware dataset AdwareZoo, which comprises 15,996 adware samples across 118 distinct adware families collected from both security reports and app repositories, providing a robust foundation for in-depth analysis and future research. We identify adware family payloads by isolating packages from adware samples for VirusTotal rescanning. Our analysis unveils critical insights into the adware ecosystem, highlighting distinctive patterns in family naming conventions and exposing the unexpected classification of widely-used ad networks as adware. We identify diverse payload location strategies, with a notable finding that over 30% of adware families employ payloads beyond conventional Java/Kotlin code. Furthermore, we reveal several evasion techniques utilized by ad-ware, including package name obfuscation and dynamic payload loading. This research not only offers a robust foundation for understanding and combating adware but also highlights the pressing need for increased scrutiny of mobile advertising practices. © 2024 ACM.

关键词： Android malware

来源：评论

学校读者我要写书评

暂无评论

Generating Targeted Universal Adversarial Perturbation against Automatic Speech Recognition via Phoneme Tailoring

Generating Targeted Universal Adversarial Perturbation again...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Zhang, Yujun Chen, Yanqu Wang, Jiakai Hu, Jin Tao, Renshuai Liu, Xianglong State Key Laboratory of Complex & Critical Software Environment Beihang University China School of Computer Science and Engineering Beihang University China College of Computer Science Beijing University of Technology China Zhongguancun Laboratory China School of Computer and Information Technology Beijing Jiaotong University China Institute of Data Space Hefei Comprehensive National Science Center China

ISBN: (纸本)9798350368741

There is a growing concern about adversarial attacks against automatic speech recognition (ASR) systems. Although research into targeted universal adversarial examples (AEs) has progressed, current methods are constrained by inefficient exploitation of audio features, demonstrating insufficient attack ability and robustness in the physical world. To solve this problem, we propose a phoneme-tailored attack (PTA) to improve the quality of the generated AEs. Specifically, to improve attack ability, we propose a Diverse Audio Composition Enrichment method, which enhances the utilization of audio features through phoneme-level slicing and recombination. To adapt AEs to complex environments, we propose a Natural Noise Pattern Guidance method to align AEs with natural noise patterns to improve their robustness. Experiments show that our method achieves an average accuracy of more than 72.34% and 98% with and without a norm constraint, and also demonstrates excellent performance in terms of generalization across datasets and resilience to MP3 compression. © 2025 IEEE.

关键词： Automatic speech recognition natural noise pattern phoneme-level targeted universal adversarial examples

来源：评论

学校读者我要写书评

暂无评论

Owl: Differential-Based Side-Channel Leakage Detection for CUDA Applications

Owl: Differential-Based Side-Channel Leakage Detection for C...

引用

International Conference on Dependable systems and Networks (DSN)

作者： Yu Zhao Wenjie Xue Weijie Chen Weizhong Qiang Deqing Zou Hai Jin National Engineering Research Center for Big Data Technology and System Hubei Key Laboratory of Distributed System Security Hubei Engineering Research Center on Big Data Security School of Cyber Science and Engineering Huazhong University of Science and Technology Wuhan China Jinyinhu Laboratory Wuhan China 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

ISBN: (数字)9798350341058

ISBN: (纸本)9798350341065

Over the past decade, various methods for detecting side-channel leakage have been proposed and proven to be effective against CPU side-channel attacks. These methods are valuable in assisting developers to identify and patch side-channel vulnerabilities. Nevertheless, recent research has revealed the feasibility of exploiting side-channel vulnerabilities to steal sensitive information from GPU applications, which are beyond the reach of previous side-channel detection methods. Therefore, in this paper, we conduct an in-depth examination of various GPU features and present Owl, a novel side-channel detection tool targeting CUDA applications on NVIDIA GPUs. Owl is designed to detect and locate side-channel leakage in various types of CUDA applications. When tracking the execution of CUDA applications, we design a hierarchical tracing scheme and extend the A-DCFG (Attributed Dynamic Control Flow Graph) to address the massively parallel execution in CUDA, ensuring Owl's detection scalability. After completing the initial assessment and filtering, we conduct statistical tests on the differences in program traces to determine whether they are indeed caused by input variations, subsequently facilitating the positioning of side-channel leaks. We evaluate Owl's capability to detect side-channel leaks by testing it on Libgpucrypto, PyTorch, and nvJPEG. Meanwhile, we verify that our solution effectively handles a large number of threads. Owl has successfully identified hundreds of leaks within these applications. To the best of our knowledge, we are the first to implement side-channel leakage detection for general CUDA applications.

关键词： Target tracking Statistical analysis Filtering Scalability Graphics processing units Side-channel attacks Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Intersecting-boundary-sensitive fingerprinting for tampering detection of DNN models 24

Intersecting-boundary-sensitive fingerprinting for tampering...

引用

Proceedings of the 41st International Conference on Machine Learning

作者： Xiaofan Bai Chaoxiang He Xiaojing Ma Bin Benjamin Zhu Hai Jin School of Cyber Science and Engineering Huazhong University of Science and Technology National Engineering Research Center for Big Data Technology and System and Services Computing Technology and System Lab and Hubei Engineering Research Center on Big Data Security and Hubei Key Laboratory of Distributed System Security Microsoft School of Computer Science and Technology Huazhong University of Science and Technology and National Engineering Research Center for Big Data Technology and System and Services Computing Technology and System Lab and Cluster and Grid Computing Lab.

Cloud-based AI services offer numerous benefits but also introduce vulnerabilities, allowing for tampering with deployed DNN models, ranging from injecting malicious behaviors to reducing computing resources. Fingerprint samples are generated to query models to detect such tampering. In this paper, we present Intersecting-Boundary-Sensitive Fingerprinting (IBSF), a novel method for black-box integrity verification of DNN models using only top-1 labels. Recognizing that tampering with a model alters its decision boundary, IBSF crafts fingerprint samples from normal samples by maximizing the partial Shannon entropy of a selected subset of categories to position the fingerprint samples near decision boundaries where the categories in the subset intersect. These fingerprint samples are almost indistinguishable from their source samples. We theoretically establish and confirm experimentally that these fingerprint samples' expected sensitivity to tampering increases with the cardinality of the subset. Extensive evaluation demonstrates that IBSF surpasses existing state-of-the-art fingerprinting methods, particularly with larger subset cardinality, establishing its state-of-the-art performance in black-box tampering detection using only top-1 labels. The IBSF code is available at: https://***/CGCL-codes/IBSF.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Understanding the Low Predictability of the 2015/16 El Niño Event Based on a Deep Learning Model

引用

Advances in Atmospheric Sciences 2024年第7期41卷 1313-1325页

作者： Tingyu WANG Ping HUANG Xianke YANG Center for Monsoon System Research Institute of Atmospheric PhysicsChinese Academy of SciencesBeijing 100029China National Space Science Center Chinese Academy of SciencesBeijing 100190China State Key Laboratory of Space Weather National Space Science CenterChinese Academy of SciencesBeijing 100190China College of Plant Science and Technology Huazhong Agricultural UniversityWuhan 430070China

The 2015/16 El Niño event ranks among the top three of the last 100 years in terms of intensity,but most dynamical models had a relatively low prediction skill for this event before the summer ***,the attribution of this particular event can help us to understand the cause of super El Niño–Southern Oscillation events and how to forecast them *** present study applies attribute methods based on a deep learning model to study the key factors related to the formation of this event.A deep learning model is trained using historical simulations from 21 CMIP6 models to predict the Niño-3.4 *** integrated gradient method is then used to identify the key signals in the North Pacific that determine the evolution of the Niño-3.4 *** crucial signals are then masked in the initial conditions to verify their roles in the *** addition to confirming the key signals inducing the super El Niño event revealed in previous attribution studies,we identify the combined contribution of the tropical North Atlantic and the South Pacific oceans to the evolution and intensity of this event,emphasizing the crucial role of the interactions among them and the North *** approach is also applied to other El Niño events,revealing several new precursor *** study suggests that the deep learning method is useful in attributing the key factors inducing extreme tropical climate events.

关键词： ENSO attribution deep learning ENSO prediction extreme El Niño

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：