检索结果-内蒙古大学图书馆

32nd ACM International Conference on Multimedia, MM 2024

作者： Wang, Yuran Wan, Zhijing Qiu, Yansheng Wang, Zheng National Engineering Research Center for Multimedia Software Hubei Key Laboratory of Multimedia and Network Communication Engineering Institute of Artificial Intelligence School of Computer Science Wuhan University Wuhan China

ISBN: (纸本)9798400706868

In the realm of medical image analysis, self-supervised learning (SSL) techniques have emerged to alleviate labeling demands, while still facing the challenge of training data scarcity owing to escalating resource requirements and privacy constraints. Numerous efforts employ generative models to generate high-fidelity, unlabeled 3D volumes across diverse modalities and anatomical regions. However, the intricate and indistinguishable anatomical structures within the abdomen pose a unique challenge to abdominal CT volume generation compared to other anatomical regions. To address the overlooked challenge, we introduce the Locality-Aware Diffusion (Lad), a novel method tailored for exquisite 3D abdominal CT volume generation. We design a locality loss to refine crucial anatomical regions and devise a condition extractor to integrate abdominal priori into generation, thereby enabling the generation of large quantities of high-quality abdominal CT volumes essential for SSL tasks without the need for additional data such as labels or radiology reports. Volumes generated through our method demonstrate remarkable fidelity in reproducing abdominal structures, achieving a decrease in FID score from 0.0034 to 0.0002 on AbdomenCT-1K dataset, closely mirroring authentic data and surpassing current methods. Extensive experiments demonstrate the effectiveness of our method in self-supervised organ segmentation tasks, resulting in an improvement in mean Dice scores on two abdominal datasets effectively. These results underscore the potential of synthetic data to advance self-supervised learning in medical image analysis. © 2024 ACM.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

A Theoretical Analysis of Backdoor Poisoning Attacks in Convolutional Neural networks 41

A Theoretical Analysis of Backdoor Poisoning Attacks in Conv...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Li, Boqi Liu, Weiwei School of Computer Science National Engineering Research Center for Multimedia Software Institute of Artificial Intelligence Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan University Wuhan China

The rising threat of backdoor poisoning attacks (BPAs) on Deep Neural networks (DNNs) has become a significant concern in recent years. In such attacks, the adversaries strategically target a specific class and generate a poisoned training set. The neural network (NN), well-trained on the poisoned training set, is able to predict any input with the trigger pattern as the targeted label, while maintaining accurate outputs for clean inputs. However, why the BPAs work remains less explored. To fill this gap, we employ a dirty-label attack and conduct a detailed analysis of BPAs in a two-layer convolutional neural network. We provide theoretical insights and results on the effectiveness of BPAs. Our experimental results on two real-world datasets validate our theoretical findings. Copyright 2024 by the author(s)

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Never Lost Keys:A Novel Key Generation Scheme Based on Motor Imagery EEG in End-Edge-Cloud System

引用

China Communications 2022年第7期19卷 172-184页

作者： Yichuan Wang Dan Wu Xiaoxue Liu Xinhong Hei School of computer science and Engineering Xi’an University of TechnologyXi’an 710048China Shaanxi Key Laboratory for Network Computing and Security Technology Xi’an 710048China

Biometric key is generated from the user’s unique biometric features,and can effectively solve the security problems in ***,the current prevailing biometric key generation techniques such as fingerprint recognition and facial recognition are poor in randomness and can be forged *** to the characteristics of Electroencephalographic(EEG)signals such as the randomness,nonlinear and non-stationary etc.,it can significantly avoid these *** paper proposes a novel method to generate keys based on EEG signals with end-edgecloud collaboration *** sensors to measure motor imagery EEG data,the key is generated via pre-processing,feature extraction and *** show the total time consumption of the key generation process is about *** scheme is practical and feasible,which provides a research route to generate biometric keys using EEG data.

关键词： EEG biometric key generation end-edgecloud system information security

来源：评论

学校读者我要写书评

暂无评论

Generalization Analysis for Label-Specific Representation Learning 38

Generalization Analysis for Label-Specific Representation Le...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Zhang, Yi-Fan Zhang, Min-Ling School of Cyber Science and Engineering Southeast University Nanjing210096 China School of Computer Science and Engineering Southeast University Nanjing210096 China Key Laboratory of Computer Network and Information Integration Southeast University Ministry of Education China

Label-specific representation learning (LSRL), i.e., constructing the representation with specific discriminative properties for each class label, is an effective strategy to improve the performance of multi-label learning. However, the generalization analysis of LSRL is still in its infancy. The existing theory bounds for multilabel learning, which preserve the coupling among different components, are invalid for LSRL. In an attempt to overcome this challenge and make up for the gap in the generalization theory of LSRL, we develop a novel vector-contraction inequality and derive the generalization bound for general function class of LSRL with a weaker dependency on the number of labels than the state of the art. In addition, we derive generalization bounds for typical LSRL methods, and these theoretical results reveal the impact of different label-specific representations on generalization analysis. The mild bounds without strong assumptions explain the good generalization ability of LSRL. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Exploring MEC Server Strategy in Blockchain networks: Mining for Mobile Users or for Self

引用

IEEE Transactions on Mobile Computing 2025年第7期24卷 6030-6044页

作者： Ling, Xintong Jiang, Rui Cao, Weihang Chen, Mingkai Wang, Jiaheng Ding, Zhi Gao, Xiqi Southeast University National Mobile Communications Research Laboratory Nanjing210096 China Purple Mountain Laboratories Nanjing210023 China Nanjing University of Posts and Telecommunications Key Laboratory of Broadband Wireless Communication and Sensor Network Technology Nanjing210003 China University of Maryland Department of Electrical and Computer Engineering College ParkMD20742 United States Southeast University School of Cyber Science and Engineering Nanjing210096 China University of California Department of Electrical and Computer Engineering DavisCA95616 United States

Blockchain-based decentralized applications (DApps) offer enhanced security and decentralization features;nevertheless, their maintenance demands substantial computational resources and poses challenges for deployment in mobile networks. To address this, a number of studies have explored offloading blockchain mining tasks from mobile users to mobile edge computing (MEC) servers. However, the existing literature overlooks the fact that MEC servers can not only mine for mobile users but also mine for themselves, potentially explaining why MEC mining offloading has not gained broad acceptance within the industry. In this work, we exploit a more practical case and rethink the question of whether MEC servers lease computing power to mobile users by taking into account that MEC servers can mine for themselves. We establish a game model and apply backward induction to analytically characterize Nash equilibria for mining strategies adopted by MEC and mobile users. Our findings suggest that, if MEC can mine for self, MEC would mine for mobile users only under specific conditions where mobile users possess superior information gathering capability (at least better than the MEC server) or the whole blockchain system exhibits significant network value. We further provide a series of simulations to verify our conclusion and illustrate the impact of network parameters on the strategies of both sides. © 2002-2012 IEEE.

关键词： Mobile edge computing

来源：评论

学校读者我要写书评

暂无评论

SQAT-LD: SPeech Quality Assessment Transformer Utilizing Listener Dependent Modeling for Zero-Shot Out-of-Domain MOS Prediction

SQAT-LD: SPeech Quality Assessment Transformer Utilizing Lis...

引用

2023 IEEE Automatic Speech Recognition and Understanding Workshop, ASRU 2023

作者： Shen, Kailai Yan, Diqun Dong, Li Ren, Ying Wu, Xiaoxun Hu, Jing Ningbo University Faculty of Electrical Engineering and Computer Science China Zhejiang Key Laboratory of Mobile Network Application Technology China

ISBN: (纸本)9798350306897

In this paper, we propose the speech quality assessment transformer utilizing listener dependent modeling (SQAT-LD) mean opinion score (MOS) prediction system, which was submitted to the 2023 VoiceMOS Challenge. The system is based on a combination of self-supervised learning (SSL) models and listener-dependent modeling. Due to this challenge's emphasis on real-world and challenging zero-shot out-of-domain MOS prediction in three different voice evaluation scenarios, we specifically designed a two-branch module to predict scores and weights for each frame, aiming to achieve better generalization. In the challenge, our system achieved fourth place in Track 1a, second place in Track 1b and first place in Track 2. Additionally, we conducted an ablation study to investigate the effectiveness of our proposed method. © 2023 IEEE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

IGED:Towards Intelligent DDoS Detection Model Using Improved Generalized Entropy and DNN

引用

computers, Materials & Continua 2024年第8期80卷 1851-1866页

作者： Yanhua Liu Yuting Han HuiChen Baokang Zhao XiaofengWang Ximeng Liu College of Computer and Data Science Fuzhou UniversityFuzhou350108China Engineering Research Center of Big Data Intelligence Ministry of EducationFuzhou350108China Fujian Key Laboratory of Network Computing and Intelligent Information Processing Fuzhou UniversityFuzhou350108China College of Computer National University of Defense TechnologyChangsha410073China

As the scale of the networks continually expands,the detection of distributed denial of service(DDoS)attacks has become increasingly *** propose an intelligent detection model named IGED by using improved generalized entropy and deep neural network(DNN).The initial detection is based on improved generalized entropy to filter out as much normal traffic as possible,thereby reducing data *** the fine detection is based on DNN to perform precise DDoS detection on the filtered suspicious traffic,enhancing the neural network’s generalization *** results show that the proposed method can efficiently distinguish normal traffic from DDoS *** with the benchmark methods,our method reaches 99.9%on low-rate DDoS(LDDoS),flooded DDoS and CICDDoS2019 datasets in terms of both accuracy and efficiency in identifying attack flows while reducing the time by 17%,31%and 8%.

关键词： DDoS real-time improved generalized entropy DNN

来源：评论

学校读者我要写书评

暂无评论

Bit level image encryption algorithm based on hyperchaotic system

引用

Optoelectronics Letters 2023年第3期19卷 186-192页

作者： MAN Zhenlong ZHANG Yue ZHOU Ying LU Xiaoli WANG Zhaoquan School of Electronic and Information Engineering Liaoning Technical UniversityHuludao 125100China Advanced Interactive Technology and Application Laboratory Computer Network Information CenterChinese Academy of SciencesBeijing 100190China

Because chaotic systems are unpredictable,ergodic and sensitive to initial values and parameters,they are often used in the field of *** avoid the bad randomness of the random key generated by the low dimensional chaotic map system,a 5-dimensional multi-wing hyperchaotic system is adopted in this *** key stream generated by the chaotic system is related to the plaintext *** it can effectively resist the attacks of selecting *** plaintext graph is decomposed into binary form by bit plane decomposition technique,and then these bit planes are divided into high and low *** designed control matrix is used to identify the specific scrambling mode,and each bit is permuted within and between ***,bit diffusion is used to change the pixel value of each *** analysis and numerical simulation show that the algorithm has good encryption performance for image encryption.

关键词： chaotic system algorithm

来源：评论

学校读者我要写书评

暂无评论

A Linked List Encryption Scheme for Image Steganography without Embedding

引用

computer Modeling in engineering & Sciences 2024年第10期141卷 331-352页

作者： Pengbiao Zhao Qi Zhong Jingxue Chen Xiaopei Wang Zhen Qin Erqiang Zhou Network and Data Security Key Laboratory of Sichuan Province University of Electronic Science and Technology of ChinaChengdu610054China Faculty of Data Science City University of MacaoMacao999078China Department of Computer Science and Engineering University of CaliforniaRiversideCA 92521USA

Information steganography has received more and more attention from scholars nowadays,especially in the area of image steganography,which uses image content to transmit information and makes the existence of secret information *** enhance concealment and security,the Steganography without Embedding(SWE)method has proven effective in avoiding image distortion resulting from cover *** this paper,a novel encrypted communication scheme for image SWE is *** reconstructs the image into a multi-linked list structure consisting of numerous nodes,where each pixel is transformed into a single node with data and pointer *** employing a special addressing algorithm,the optimal linked list corresponding to the secret information can be *** receiver can restore the secretmessage fromthe received image using only the list header position *** scheme is based on the concept of coverless steganography,eliminating the need for any modifications to the cover *** boasts high concealment and security,along with a complete message restoration rate,making it resistant to ***,this paper proposes linked-list construction schemeswithin theproposedframework,which caneffectively resist a variety of attacks,includingnoise attacks and image compression,demonstrating a certain degree of *** validate the proposed framework,practical tests and comparisons are conducted using multiple *** results affirm the framework’s commendable performance in terms of message reduction rate,hidden writing capacity,and robustness against diverse attacks.

关键词： Steganography encryption steganography without embedding coverless steganography

来源：评论

学校读者我要写书评

暂无评论

A Deep Understanding Video Q&A System for Film Education in Acting Department

A Deep Understanding Video Q&A System for Film Education in ...

引用

2023 International Conference on Intelligent Education and Intelligent Research, IEIR 2023

作者： Wu, Zhengqian Li, Ruizhe Guo, Jiahao Wang, Zhongyuan Liang, Chao Hubei Key Laboratory of Multimedia and Network Communication Engineering School of Computer Science China National Engineering Research Center for Multimedia Software School of Computer Science China Wuhan University School of Computer Science China

ISBN: (纸本)9798350342895

Recently, advancements in artificial intelligence technology have greatly influenced the field of education, particularly in the area of intelligent homework assistance. However, current approaches are primarily designed for procedural and logical tasks and often lack comprehension abilities. This limitation is particularly evident when it comes to multi-hop and continuous tasks. To address this challenge, the integration of Large Language Model (LLM) has significantly enhanced the capability of AI systems to handle multi-hop and highly interconnected inputs. In this study, we focus on the learning needs of students in Acting Department, specifically their study of movies and the significance of classic movie videos in their learning process. However, assessing deep comprehension of classic movies poses its own challenges. To overcome these challenges, we develop a quiz system utilizing Knowledge Graphs (KG) and LLM to facilitate a deeper understanding of classic films. The generation of video quiz pairs is achieved through the use of Automatic Speech Recognition (ASR) technology, which leverages movie subtitles for question generation. For answering these questions, we employ techniques KG and LLM to process questions and retrieve corresponding answers. The proposed method achieves good performance in Deep Video Understanding (DVU) task of NIST TRECVID, demonstrating its effectiveness. © 2023 IEEE.

关键词： Motion pictures

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：