检索结果-内蒙古大学图书馆

IEEE International Conference on Big Data and Cloud Computing (BdCloud)

作者： Chunming Liu Daniyal M Alghazzawi Li Cheng Gaoyang Liu Chen Wang Cheng Zeng Yang Yang Key Laboratory of Intelligent Sensing System and Security (Ministry of Education) School of Artificial Intelligence Hubei University Wuhan China King Abdulaziz University Jeddah Saudi Arabia Huazhong University of Science and Technology Wuhan China

Federated Learning (FL) is a promising paradigm that leverages distributed data sources to train machine learning models, thereby offering significant privacy advantages. However, the inherent statistical heterogeneity among clients, characterized by distinct data distributions and varying model performance, poses a substantial challenge. Such heterogeneity adversely affects both the convergence and overall performance of the global model within the FL framework. To address this issue, we introduce a novel FL algorithm, FedVa, which considers both the local data volume and model accuracy to determine client contributions, assigning respective weights. This consideration facilitates greater engagement from key clients during the aggregation process, thereby enhancing their influence on the global model refinement. Experimental results indicate that FedVa surpasses prevailing methods in managing statistical heterogeneity and enhancing global model accuracy, without incurring additional communication costs. Specifically, FedVa achieves a convergence speed 3.9 times faster than FedAvg on the MNIST dataset, and on the CIFAR-10 dataset, it results in a 2.3% increase in final model accuracy compared to FedAvg. Our code is publicly available at https://***/ChunmingLiu23/FedVa.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Contrastive Learning-Based Generic Audio-to-Lyrics Alignment

Contrastive Learning-Based Generic Audio-to-Lyrics Alignment

引用

IEEE International Conference on Big Data and Cloud Computing (BdCloud)

作者： Zhenkun Jin Shuai Huang Xin Nie Xinlei Zhou Yuanyuan Yi Gefei Zhou Wuhan Business UIniversity Wuhan China Wuhan Research Institute of Posts and Telecommunications Wuhan China Hubei Key Laboratory of Distributed System Security Wuhan China Hubei Engineering Research Center on BigData Security Wuhan China School of Cyber Science and Engineering Huazhong University of Science and Technology Wuhan China School of Journalism and Information Communication Huazhong University of Science and Technology Wuhan China School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China

In this research, we tackled the complex task of aligning audio and lyrics content automatically. This task entails the precise matching of lyrics with the corresponding audio segments in songs, necessitating the coordination of two distinct modalities: text and audio. To surmount this challenge, we employed a contrastive learning approach that generates cross-modal embeddings capable of establishing connections between the audio and text domains. Furthermore, we adopted a hierarchical and cascading design. Initially, the system predicts synchronization at the sentence level and subsequently refines the alignment at the word level. This design equips the system to handle extended sequences effectively, as cross-correlations leverage secondary memory that corresponds to sequence length. Through this innovative approach, we developed a trainable end-to-end system that can harness weakly annotated training data, concurrently learning a robust text model optimized for alignment tasks. Our experiments demonstrated substantial enhancements in average alignment accuracy when utilizing our proposed system, underscoring its robustness compared to previous state-of-the-art models.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Theoretical study of the interaction of SF6 molecule on Ti3C2Tx surfaces

SSRN

引用

SSRN 2022年

作者： Zeng, Fuping Guo, Xinnuo Feng, Xiaoxuan Cai, Rijian Yao, Qiang Zhang, Shiling Tang, Ju School of Electrical Engineering and Automation Wuhan University Wuhan430072 China Hubei Key Laboratory of Power Equipment & System Security for Integrated Energy Resources Wuhan430072 China State Grid Chongqing Electric Power Company Electric Power Research Institute Chongqing401123 China State Key Laboratory of Power Transmission Equipment & System Security and New Technology Chongqing University Chongqing400044 China

At present, searching for efficient methods to degrade SF6 is a worldwide hotspot. Due to the abundant active groups on the surface, MXene is regarded as an ideal catalyst support. In this paper, the potential active sites of Ti3C2Tx (T=OH, F, O) for SF6 catalytic degradation were studied using DFT method. The results show the adsorption of SF6 on Ti3C2(OH)2 is significantly stronger than that on Ti3C2F2 and Ti3C2O2. Among typical adsorption sites on Ti3C2(OH)2, SF6 exhibited the strongest interaction at OH-1 site. Complex electron orbital interactions occurred between SF6 and Ti3C2(OH)2. The adsorption energy reached -6.739eV and about 1.414e transferred from Ti3C2(OH)2 to SF6. Structure of SF6 changed obviously and chemical adsorption occurred in this process. However, on Ti3C2F2 and Ti3C2O2, adsorption energy and electron transfer were small and the structure had little change, which was mainly physical adsorption. Compared with common catalysts such as Ag, α‑Al2O3 and BaTiO3, Ti3C2(OH)2 showed better catalytic activity for SF6 degradation. By strong adsorption of the active site on the surface, the degradation products were separated from each other, which weakened SF6 recombination and promoted the further degradation. The results provide theoretical support for MXene acting as a catalyst to efficiently degrade SF6 © 2022, The Authors. All rights reserved.

关键词： Sulfur hexafluoride

来源：评论

学校读者我要写书评

暂无评论

Downstream-agnostic Adversarial Examples

arXiv

引用

arXiv 2023年

作者： Zhou, Ziqi Hu, Shengshan Zhao, Ruizhi Wang, Qian Zhang, Leo Yu Hou, Junhui 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 Cyber Science and Engineering Wuhan University China School of Information and Communication Technology Griffith University Australia Department of Computer Science City University of Hong Kong Hong Kong 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

Self-supervised learning usually uses a large amount of unlabeled data to pre-train an encoder which can be used as a general-purpose feature extractor, such that downstream users only need to perform fine-tuning operations to enjoy the benefit of "large model". Despite this promising prospect, the security of pre-trained encoder has not been thoroughly investigated yet, especially when the pre-trained encoder is publicly available for commercial use. In this paper, we propose AdvEncoder, the first framework for generating downstream-agnostic universal adversarial examples based on the pre-trained encoder. AdvEncoder aims to construct a universal adversarial perturbation or patch for a set of natural images that can fool all the downstream tasks inheriting the victim pre-trained encoder. Unlike traditional adversarial example works, the pre-trained encoder only outputs feature vectors rather than classification labels. Therefore, we first exploit the high frequency component information of the image to guide the generation of adversarial examples. Then we design a generative attack framework to construct adversarial perturbations/patches by learning the distribution of the attack surrogate dataset to improve their attack success rates and transferability. Our results show that an attacker can successfully attack downstream tasks without knowing either the pre-training dataset or the downstream dataset. We also tailor four defenses for pre-trained encoders, the results of which further prove the attack ability of AdvEncoder. Our codes are available at: https://github. com/CGCL-codes/AdvEncoder. Copyright © 2023, The Authors. All rights reserved.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Detecting Backdoors During the Inference Stage Based on Corruption Robustness Consistency

arXiv

引用

arXiv 2023年

作者： Liu, Xiaogeng Li, Minghui Wang, Haoyu Hu, Shengshan Ye, Dengpan Jin, Hai Wu, Libing Xiao, Chaowei 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 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 School of Cyber Science and Engineering Wuhan University China Arizona State University United States

Deep neural networks are proven to be vulnerable to backdoor attacks. Detecting the trigger samples during the inference stage, i.e., the test-time trigger sample detection, can prevent the backdoor from being triggered. However, existing detection methods often require the defenders to have high accessibility to victim models, extra clean data, or knowledge about the appearance of backdoor triggers, limiting their practicality. In this paper, we propose the test-time corruption robustness consistency evaluation (TeCo)1, a novel test-time trigger sample detection method that only needs the hard-label outputs of the victim models without any extra information. Our journey begins with the intriguing observation that the backdoor-infected models have similar performance across different image corruptions for the clean images, but perform discrepantly for the trigger samples. Based on this phenomenon, we design TeCo to evaluate test-time robustness consistency by calculating the deviation of severity that leads to predictions' transition across different corruptions. Extensive experiments demonstrate that compared with state-of-the-art defenses, which even require either certain information about the trigger types or accessibility of clean data, TeCo outperforms them on different backdoor attacks, datasets, and model architectures, enjoying a higher AUROC by 10% and 5 times of stability. Copyright © 2023, The Authors. All rights reserved.

关键词： Crime

来源：评论

学校读者我要写书评

暂无评论

PointCA: Evaluating the Robustness of 3D Point Cloud Completion Models Against Adversarial Examples

arXiv

引用

arXiv 2022年

作者： Hu, Shengshan Zhang, Junwei Liu, Wei Hou, Junhui Li, Minghui Zhang, Leo Yu Jin, Hai Sun, Lichao 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 Cluster and Grid Computing Lab China National Engineering Research Center for Big Data Technology and System Hubei Engineering Research Center on Big Data Security China Hubei Key Laboratory of Distributed System Security China Services Computing Technology and System Lab China City University of Hong Kong Hong Kong Deakin University Australia Lehigh University United States

Point cloud completion, as the upstream procedure of 3D recognition and segmentation, has become an essential part of many tasks such as navigation and scene understanding. While various point cloud completion models have demonstrated their powerful capabilities, their robustness against adversarial attacks, which have been proven to be fatally malicious towards deep neural networks, remains unknown. In addition, existing attack approaches towards point cloud classifiers cannot be applied to the completion models due to different output forms and attack purposes. In order to evaluate the robustness of the completion models, we propose PointCA, the first adversarial attack against 3D point cloud completion models. PointCA can generate adversarial point clouds that maintain high similarity with the original ones, while being completed as another object with totally different semantic information. Specifically, we minimize the representation discrepancy between the adversarial example and the target point set to jointly explore the adversarial point clouds in the geometry space and the feature space. Furthermore, to launch a stealthier attack, we innovatively employ the neighbourhood density information to tailor the perturbation constraint, leading to geometry-aware and distribution-adaptive modification for each point. Extensive experiments against different premier point cloud completion networks show that PointCA can cause a performance degradation from 77.9% to 16.7%, with the structure chamfer distance kept below 0.01. We conclude that existing completion models are severely vulnerable to adversarial examples, and state-of-the-art defenses for point cloud classification will be partially invalid when applied to incomplete and uneven point cloud data. Copyright © 2022, The Authors. 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

Assessment on Transient Reactive Power Capability of a Grid-Connected Synchronous Condenser

Assessment on Transient Reactive Power Capability of a Grid-...

引用

International Conference on Electrical Machines and systems (ICEMS)

作者： Jianfu Li Hui Li Wang Xiao Bin Yuan Youbin Zhou Tao Wang Guanghou Zhou State Key Laboratory of Power Transmission Equipment & System Security and New Technology Chongqing University Chongqing China State Grid Hubei Electric Power Company Electric Power Research Institute Hubei China Dongfang Electric Machinery Co. Ltd Deyang China

ISBN: (纸本)9781728133997

New generation of large-capacity synchronous condensers (SC) is used in ultrahigh-voltage direct current (UHVDC) engineering. It aims to improve the commutation failure of DC system and suppress the transient over-voltage of the system by using its dynamic reactive characteristics. Therefore, it is necessary to accurately evaluate the transient reactive output capability of the SC. Firstly, based on the mathematical model of SC, the transient reactive power expression in the grid fault is deduced. Secondly, the saturation of the SC parameters, the steady-state reactive power, and the degree of grid voltage change are analyzed. Finally, the SC simulation model is established based on the actual SC electromagnetic parameters, and its transient reactive output characteristics are simulated and compared with the theoretical analysis results. The results show that the transient reactive power output of the SC will increase when parameters become saturated. The output transient reactive power increment of the SC increases with the increase of the surge degree of power grid voltage. The output transient reactive power increment of the SC increases and then decreases with the drop degree of power grid voltage. It can provide more effective reactive power support for the system when the steady-state operation of the SC is in phase advance and when the power grid fails.

关键词： Transient analysis Reactive power Mathematical model Windings Steady-state Stator windings

来源：评论

学校读者我要写书评

暂无评论

Modulation of Switching Current Density in T-Type Magnetic Structure Through Magnetic Anisotropy

SSRN

引用

SSRN 2023年

作者： Li, Peizhi Wu, Chuangwen Zhou, Xiangqing Ren, Chuantong Cao, Cuimei Su, Yurong Chen, Shiwei Liang, Shiheng School of Physics Hubei University Wuhan430062 China School of Integrated Circuits Wuhan National Laboratory for Optoelectronics Huazhong University of Science and Technology Wuhan430074 China Key Laboratory for Intelligent Sensing System and Security of Ministry of Education Wuhan430062 China

We investigate magnetic field driven domain wall motion and spin-orbit torque (SOT) induced perpendicular magnetization switching of the ferrimagnetic layer in a T-type magnetic heterojunction with the structure of Pt/Co/Ta/Co81Tb19/Ta, where the bottom Co layer has in-plane magnetic anisotropy and the top Co81Tb19 layer has perpendicular magnetic anisotropy. SOT-induced perpendicular magnetization switching of the Co81Tb19 layer is observed. A lower critical switching current density was obtained at the optimized thickness (tCo = 1.2 nm), resulting in a 37% reduction in critical switching current density. Meanwhile, the research reveals a strong correlation between critical switching current density and magnetic anisotropy field, indicating that tuning the magnetic anisotropy is an efficient way to modulate the switching current density in a T-type magnetic structure. These results provide a new perspective on SOT manipulation and will be useful for the design of SOT-based devices. © 2023, The Authors. All rights reserved.

关键词： Binary alloys

来源：评论

学校读者我要写书评

暂无评论

Experimental and simulation analysis of the influence of moisture on dielectric and breakdown properties of insulation pressboard used in power transformer 3

Experimental and simulation analysis of the influence of moi...

引用

2020 3rd International Conference on Mechatronics and Computer Technology Engineering, MCTE 2020

作者： Wen, Hao Cheng, Lin Jiang, Yi Zhao, Zhigang Ye, Wenyu Hao, Jian Nanjing Nari Group Corp. State Grid Electric Power Research Institute Nanjing Jiangsu211000 China Wuhan Nari Co Ltd. State Grid Electric Power Research Institute Wuhan Hubei430074 China China Electric Power Research Institute Beijing100192 China State Key Laboratory of Power Transmission Equipment and System Security and New Technology Chongqing University Chongqing400044 China

The influence of water molecules on cellulose has attracted much attention. The influence of water molecules on the dielectric property and breakdown property of cellulose requires in-depth research. In this paper, firstly, the effect of water on the dielectric spectrum in the frequency domain dielectric spectrum and the breakdown voltage of cellulose were studied. Then the parameters such as free volume, hydrogen bond, energy level and state density of cellulose under different moisture content were analysed by molecular simulation. The changes of dielectric properties and breakdown properties of cellulose under different water content were studied and summarized. The variation mechanism of dielectric property and breakdown property of different moisture content was explained by molecular simulation. © Published under licence by IOP Publishing Ltd.

关键词： Cellulose

来源：评论

学校读者我要写书评

暂无评论

Tensor Dynamic Fusion Based Modality-Imbalanced Multimodal Federated Learning in Mobile Edge Computing for Consumer Applications

引用

IEEE Transactions on Consumer Electronics 2024年

作者： Nie, Xin Yi, Lingzhi Yang, Laurence T. Deng, Xianjun Fan, Fulan Zhang, Hongtao Huazhong University of Science and Technology Hubei Key Laboratory of Distributed System Security Hubei EngineeringResearch Center on Big Data Security School of Cyber Science and Engineering Wuhan430074 China Zhongnan University of Economics and Law School of Information and Safety Engineering Wuhan430074 China Huazhong University of Science and Technology School of Computer Science and Technology Wuhan China Zhengzhou University School of Computer and Artificial Intelligence Zhengzhou China St. Francis Xavier University Department of Computer Science Antigonish Canada South-Central Minzu University School of Education Wuhan430074 China

Mobile edge computing extends the capabilities of computation and storage from the cloud to the edge, enabling local data storage and processing. This computing paradigm complements federated learning (FL), as it satisfies the need for privacy protection while providing efficiency and quick response. In consumer applications, multimodal data is experiencing explosive growth. Therefore, multimodal FL has emerged as a crucial research area. However, applying traditional FL methods directly to multimodal data poses certain challenges. Additionally, issues like sensor failures result in modality-imbalanced data among clients, i.e., the number and type of data modalities between client devices vary. Hence, it is imperative to design specialized FL frameworks and models specifically tailored for analyzing multimodal data. To address these challenges, we propose a tensor fusion-based multimodal federated learning (TenMFL). First, a tensor dynamic fusion network is designed for addressing the challenge of multimodal data fusion. This approach separates the parameter forward computation and the modal fusion computation, effectively resolving issues related to client-side modality imbalance. Second, we introduce a hybrid fusion method that combines feature fusion and decision fusion to further exploit the complementarity of multimodal data. We evaluate TenMFL on three multimodal datasets, experimental results demonstrate our superior performances. © 1975-2011 IEEE.

关键词： Federated learning

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：