检索结果-内蒙古大学图书馆

Hierarchical Reverse Games for the Resource Ecosystem in Cloud-Edge computing

学校读者我要写书评

暂无评论

IEEE Transactions on Consumer Electronics 2024年

作者： Pei, Manlin Wang, Chenyu Wang, Xia Zheng, Zhigao Huang, Jianhui Wang, Shengling Guo, Yufei School of Computer and Information Technology Beijing Jiaotong University Beijing China Department of Computer Science Georgia State University United States Beijing University of Technology School of Statistics and Data Science Faculty of Science Beijing China Institute of Artificial Intelligence School of Computer Science National Engineering Research Center for Multimedia Software Hubei Key Laboratory of Multimedia and Network Communication Engineering Hubei Luojia Laboratory Wuhan University Wuhan China Chinese Academy of Sciences Institute of Computing Technology Beijing China Beijing Normal University School of Artificial Intelligence Beijing China George Washington University Department of Statistics United States

Edge computing has emerged as a killer technology for a hyper-connected world due to its distributed architecture and customer-proximity property. Combined edge nodes with the cloud data center, a cloud-edge computing paradigm is formed, whose resource ecosystem calls for competitive pricing and optimal resource allocation. However, it is not feasible to realize global information-based optimization since some information is kept private to their owners. To solve this challenge, this paper develops the user-independent hierarchical reverse game and the user-assisted hierarchical reverse game to depict the relationship among the resource provider, the resource tenant, and the resource user. The aim is to seek the optimal strategies for players without revealing any private information. The common trait of both games is to force the players to report the optimal strategies based on their actual private information. This salient trait can produce privacy-preservation optimization since there is no requirement for the players to release their individual information directly, and the optimal strategies of other players can be deduced naturally. The difference between the user-independent hierarchical reverse game and the user-assisted one is that the game rules are stipulated entirely by the provider and the tenants (edge nodes) in the former, while some game rules can be customized according to the advice of users in the latter. This leads the user-independent hierarchical reverse game is more concise and easy to guarantee the incentive compatibility constraint, while the user-assisted hierarchical reverse game can attract more users since the game rules can adapt to their economic status, requirements, and preferences. © 1975-2011 IEEE.

关键词： Cloud platforms

DarkFed: A Data-Free Backdoor Attack in Federated Learning

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Li, Minghui Wan, Wei Ning, Yuxuan Hu, Shengshan Xue, Lulu Zhang, Leo Yu Wang, Yichen School of Software Engineering Huazhong University of Science and Technology China National Engineering Research Center for Big Data Technology and System China Services Computing Technology and System Lab China Hubei Engineering Research Center on Big Data Security China Hubei Key Laboratory of Distributed System Security China School of Cyber Science and Engineering Huazhong University of Science and Technology China School of Computer Science and Technology Huazhong University of Science and Technology China School of Information and Communication Technology Griffith University Australia

Federated learning (FL) has been demonstrated to be susceptible to backdoor attacks. However, existing academic studies on FL backdoor attacks rely on a high proportion of real clients with main task-related data, which is impractical. In the context of real-world industrial scenarios, even the simplest defense suffices to defend against the state-of-the-art attack, 3DFed. A practical FL backdoor attack remains in a nascent stage of development. To bridge this gap, we present DarkFed. Initially, we emulate a series of fake clients, thereby achieving the attacker proportion typical of academic research scenarios. Given that these emulated fake clients lack genuine training data, we further propose a data-free approach to backdoor FL. Specifically, we delve into the feasibility of injecting a backdoor using a shadow dataset. Our exploration reveals that impressive attack performance can be achieved, even when there is a substantial gap between the shadow dataset and the main task dataset. This holds true even when employing synthetic data devoid of any semantic information as the shadow dataset. Subsequently, we strategically construct a series of covert backdoor updates in an optimized manner, mimicking the properties of benign updates, to evade detection by defenses. A substantial body of empirical evidence validates the tangible effectiveness of DarkFed. Copyright © 2024, The Authors. All rights reserved.

关键词： Semantics

Why Does Little Robustness Help? A Further Step Towards Understanding Adversarial Transferability

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Zhang, Yechao Hu, Shengshan Zhang, Leo Yu Shi, Junyu Li, Minghui Liu, Xiaogeng Wan, Wei Jin, Hai School of Cyber Science and Engineering Huazhong University of Science and Technology China School of Software Engineering Huazhong University of Science and Technology China School of Information and Communication Technology Griffith University Australia School of Computer Science and Technology Huazhong University of Science and Technology China National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Hubei Key Laboratory of Distributed System Security China Hubei Engineering Research Center on Big Data Security China Cluster and Grid Computing Lab

Adversarial examples for deep neural networks (DNNs) have been shown to be transferable: examples that successfully fool one white-box surrogate model can also deceive other black-box models with different architectures. Although a bunch of empirical studies have provided guidance on generating highly transferable adversarial examples, many of these findings fail to be well explained and even lead to confusing or inconsistent advice for practical use. In this paper, we take a further step towards understanding adversarial transferability, with a particular focus on surrogate aspects. Starting from the intriguing "little robustness" phenomenon, where models adversarially trained with mildly perturbed adversarial samples can serve as better surrogates for transfer attacks, we attribute it to a trade-off between two dominant factors: model smoothness and gradient similarity. Our research focuses on their joint effects on transferability, rather than demonstrating the separate relationships alone. Through a combination of theoretical and empirical analyses, we hypothesize that the data distribution shift induced by off-manifold samples in adversarial training is the reason that impairs gradient similarity. Building on these insights, we further explore the impacts of prevalent data augmentation and gradient regularization on transferability and analyze how the trade-off manifest in various training methods, thus building a comprehensive blueprint for the regulation mechanisms behind transferability. Finally, we provide a general route for constructing superior surrogates to boost transferability, which optimizes both model smoothness and gradient similarity simultaneously, e.g., the combination of input gradient regularization and sharpness-aware minimization (SAM), validated by extensive experiments. In summary, we call for attention to the united impacts of these two factors for launching effective transfer attacks, rather than optimizing one while ignoring the other,

关键词： Deep neural networks

MLFuse: Multi-Scenario Feature Joint Learning for Multi-Modality Image Fusion

学校读者我要写书评

暂无评论

IEEE Transactions on Multimedia 2025年

作者： Lei, Jia Li, Jiawei Liu, Jinyuan Wang, Bin Zhou, Shihua Zhang, Qiang Wei, Xiaopeng Kasabov, Nikola K. Dalian University Key Laboratory of Advanced Design and Intelligent Computing Ministry of Education School of Software Engineering Dalian116622 China University of Science and Technology Beijing School of Computer and Communication Engineering Beijing100083 China Dalian University of Technology School of Mechanical Engineering Dalian116024 China Dalian University of Technology School of Computer Science and Technology Dalian116024 China Auckland University of Technology Knowledge Engineering and Discovery Research Institute Auckland1010 New Zealand Ulster University Intelligent Systems Research Center LondonderryBT52 1SA United Kingdom

Multi-modality image fusion (MMIF) entails synthesizing images with detailed textures and prominent objects. Existing methods tend to use general feature extraction to handle different fusion tasks. However, these methods have difficulty breaking fusion barriers across various modalities owing to the lack of targeted learning routes. In this work, we propose a multi-scenario feature joint learning architecture, MLFuse, that employs the commonalities of multi-modality images to deconstruct the fusion progress. Specifically, we construct a cross-modal knowledge reinforcing network that adopts a multipath calibration strategy to promote information communication between different images. In addition, two professional networks are developed to maintain the salient and textural information of fusion results. The spatial-spectral domain optimizing network can learn the vital relationship of the source image context with the help of spatial attention and spectral attention. The edge-guided learning network utilizes the convolution operations of various receptive fields to capture image texture information. The desired fusion results are obtained by aggregating the outputs from the three networks. Extensive experiments demonstrate the superiority of MLFuse for infrared-visible image fusion and medical image fusion. The excellent results of downstream tasks (i.e., object detection and semantic segmentation) further verify the high-quality fusion performance of our method. The code is publicly available at https://***/jialei-sc/MLFuse. © 1999-2012 IEEE.

关键词： Semantic Segmentation

A Four-Pronged Defense Against Byzantine Attacks in Federated Learning

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Wan, Wei Hu, Shengshan Li, Minghui Lu, Jianrong Zhang, Longling Zhang, Leo Yu Jin, Hai School of Cyber Science and Engineering Huazhong University of Science and Technology China School of Software Engineering Huazhong University of Science and Technology China School of Information and Communication Technology Griffith University Australia School of Computer Science and Technology Huazhong University of Science and Technology China National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Hubei Key Laboratory of Distributed System Security China Hubei Engineering Research Center on Big Data Security China Cluster and Grid Computing Lab

Federated learning (FL) is a nascent distributed learning paradigm to train a shared global model without violating users' privacy. FL has been shown to be vulnerable to various Byzantine attacks, where malicious participants could independently or collusively upload well-crafted updates to deteriorate the performance of the global model. However, existing defenses could only mitigate part of Byzantine attacks, without providing an all-sided shield for FL. It is difficult to simply combine them as they rely on totally contradictory assumptions. In this paper, we propose FPD, a four-pronged defense against both non-colluding and colluding Byzantine attacks. Our main idea is to utilize absolute similarity to filter updates rather than relative similarity used in existingI works. To this end, we first propose a reliable client selection strategy to prevent the majority of threats in the bud. Then we design a simple but effective score-based detection method to mitigate colluding attacks. Third, we construct an enhanced spectral-based outlier detector to accurately discard abnormal updates when the training data is not independent and identically distributed (non-IID). Finally, we design update denoising to rectify the direction of the slightly noisy but harmful updates. The four sequentially combined modules can effectively reconcile the contradiction in addressing non-colluding and colluding Byzantine attacks. Extensive experiments over three benchmark image classification datasets against four state-of-the-art Byzantine attacks demonstrate that FPD drastically outperforms existing defenses in IID and non-IID scenarios (with 30% improvement on model accuracy). © 2023, CC BY.

关键词： Classification (of information)

Accepted Influence Maximization under Linear Threshold Model on Large-Scale Social Networks

学校读者我要写书评

暂无评论

Accepted Influence Maximization under Linear Threshold Model...

IEEE International Conference on Trust, Security and Privacy in computing and communications (TrustCom)

作者： Xiaojuan Yang Jiaxing Shang Linjiang Zheng Dajiang Liu Shu Fu Baohua Qiang Key Laboratory of Dependable Service Computing in Cyber Physical Society Ministry of Education College of Computer Science Chongqing University Chongqing China School of Microelectronics and Communication Engineering Chongqing University Chongqing China Guangxi Key Laboratory of Trusted Software Guilin University of Electronic Technology Guilin China

ISBN: (纸本)9781665416597

The influence maximization (IM) problem, which aims to find $k$ most influential individuals from a social network to maximize the influence spread, has been extensively studied. Existing works all rely on the assumption that influenced individuals will definitely try to propagate the information to their neighbors through social trust. However, in real-world this assumption can be over-simplistic since trust-levels among different individuals usually exhibit high diversity. As a result, an influenced individual may choose not to further propagate the information to his neighbors. Motivated by the observation, in this paper we propose a new accepted influence maximization (AIM) problem where the influenced individuals are further divided into two subgroups, i.e., accepted and active, where only active individuals will continue to propagate the information. We prove this problem is NP-hard and the objective function is submodular, based on which a greedy algorithm is proposed with ( $1-1/e-\epsilon$ ) approximation guarantee. Considering the low computational efficiency of the greedy algorithm, we further propose a scalable path-based algorithm ALDAG. We conduct experiments on real datasets and the results demonstrate the effectiveness and efficiency of our method.

关键词： Greedy algorithms Data privacy Social networking (online) Computational modeling Heuristic algorithms Linear programming Approximation algorithms

The g-dominance Relation for Preference-Based Evolutionary Multi-Objective Optimization

学校读者我要写书评

暂无评论

The g-dominance Relation for Preference-Based Evolutionary M...

2019 IEEE Congress on Evolutionary Computation, CEC 2019

作者： Luo, Wenjian Shi, Luming Lin, Xin Coello Coello, Carlos A. Anhui Province Key Laboratory of Software Engineering in Computing and Communication University of Science and Technology of China Hefei230027 China Departmento de Sistemas UAM-Azcapotzalco Mexico City Mexico

ISBN: (纸本)9781728121536

In evolutionary multi-objective optimization, the results generated by an evolutionary algorithm usually contain an approximation, as good as possible, of the entire Pareto-optimal front. However, sometimes the number of Pareto-optimal solutions may be so large that the decision maker (DM) is incapable of manipulating or understanding them. Methods for considering only the Pareto-optimal solutions that the DM prefers indeed constitute a hot research topic in the evolutionary computation field. In this paper, we introduce a new dominance relation called hat g-dominance, which is an improved version of the g-dominance relation and can be easily implemented in traditional multi-objective evolutionary algorithms. In this work, the proposed hat g-dominance is implemented in NSGA-II. Our experimental results show the effectiveness of hat g-NSGA-II with respect to the original g-NSGA-II. © 2019 IEEE.

关键词： Multiobjective optimization

MISA: UNVEILING THE VULNERABILITIES IN SPLIT FEDERATED LEARNING

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Wan, Wei Ning, Yuxuan Hu, Shengshan Xue, Lulu Li, Minghui Zhang, Leo Yu Jin, Hai School of Cyber Science and Engineering Huazhong University of Science and Technology China School of Computer Science and Technology Huazhong University of Science and Technology China School of Software Engineering Huazhong University of Science and Technology China School of Information and Communication Technology Griffith University Australia National Engineering Research Center for Big Data Technology and System China Services Computing Technology and System Lab China Hubei Key Laboratory of Distributed System Security China Hubei Engineering Research Center on Big Data Security China Cluster and Grid Computing Lab China

Federated learning (FL) and split learning (SL) are prevailing distributed paradigms in recent years. They both enable shared global model training while keeping data localized on users' devices. The former excels in parallel execution capabilities, while the latter enjoys low dependence on edge computing resources and strong privacy protection. Split federated learning (SFL) combines the strengths of both FL and SL, making it one of the most popular distributed architectures. Furthermore, a recent study has claimed that SFL exhibits robustness against poisoning attacks, with a fivefold improvement compared to FL in terms of robustness. In this paper, we present a novel poisoning attack known as MISA. It poisons both the top and bottom models, causing a misalignment in the global model, ultimately leading to a drastic accuracy collapse. This attack unveils the vulnerabilities in SFL, challenging the conventional belief that SFL is robust against poisoning attacks. Extensive experiments demonstrate that our proposed MISA poses a significant threat to the availability of SFL, underscoring the imperative for academia and industry to accord this matter due attention. Copyright © 2023, The Authors. All rights reserved.

关键词：

A Quantum Probability Driven Framework for Joint Multi-Modal Sarcasm, Sentiment and Emotion Analysis

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Liu, Yaochen Zhang, Yazhou Song, Dawei School of Computer Science and Technology Beijing Institute of Technology Beijing100081 China School of Computing and Communications The Open University Walton Hall United Kingdom School of Nursing The Hong Kong Polytechnic University Hong Kong Ministry of Education China Artificial Intelligence Laboratory China Mobile Communication Group Tianjin Co. Ltd. China State Key Lab. for Novel Software Technology Nanjing University Nanjing China Zhengzhou University of Light Industry China

Sarcasm, sentiment, and emotion are three typical kinds of spontaneous affective responses of humans to external events and they are tightly intertwined with each other. Such events may be expressed in multiple modalities (e.g., linguistic, visual and acoustic), e.g., multi-modal conversations. Joint analysis of humans’ multi-modal sarcasm, sentiment, and emotion is an important yet challenging topic, as it is a complex cognitive process involving both cross-modality interaction and cross-affection correlation. From the probability theory perspective, cross-affection correlation also means that the judgments on sarcasm, sentiment, and emotion are incompatible. However, this exposed phenomenon cannot be sufficiently modelled by classical probability theory due to its assumption of compatibility. Neither do the existing approaches take it into consideration. In view of the recent success of quantum probability (QP) in modeling human cognition, particularly contextual incompatible decision making, we take the first step towards introducing QP into joint multi-modal sarcasm, sentiment, and emotion analysis. Specifically, we propose a QUantum probabIlity driven multi-modal sarcasm, sEntiment and emoTion analysis framework, termed QUIET. Extensive experiments on two datasets and the results show that the effectiveness and advantages of QUIET in comparison with a wide range of the state-of-the-art baselines. We also show the great potential of QP in multi-affect analysis. Copyright © 2023, The Authors. All rights reserved.

关键词： Emotion Recognition