检索结果-内蒙古大学图书馆

MEAT: MEDIAN-ENSEMBLE ADVERSARIAL TRAINING FOR IMPROVING ROBUSTNESS AND GENERALIZATION

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Hu, Zhaozhe Yin, Jia-Li Chen, Bin Lin, Luojun Chen, Bo-Hao Liu, Ximeng Fujian Province Key Laboratory of Information Security and Network Systems Fuzhou350108 China College of Computer Science and Big Data Fuzhou University Fuzhou350108 China Department of Computer Science and Engineering Yuan Ze University Taiwan

Self-ensemble adversarial training methods improve model robustness by ensembling models at different training epochs, such as model weight averaging (WA). However, previous research has shown that self-ensemble defense methods in adversarial training (AT) still suffer from robust overfitting, which severely affects the generalization performance. Empirically, in the late phases of training, the AT becomes more overfitting to the extent that the individuals for weight averaging also suffer from overfitting and produce anomalous weight values, which causes the self-ensemble model to continue to undergo robust overfitting due to the failure in removing the weight anomalies. To solve this problem, we aim to tackle the influence of outliers in the weight space in this work and propose an easy-to-operate and effective Median-Ensemble Adversarial Training (MEAT) method to solve the robust overfitting phenomenon existing in self-ensemble defense from the source by searching for the median of the historical model weights. Experimental results show that MEAT achieves the best robustness against the powerful AutoAttack and can effectively allievate the robust overfitting. We further demonstrate that most defense methods can improve robust generalization and robustness by combining with MEAT. Copyright © 2024, The Authors. All rights reserved.

关键词： Network security

Secrecy Performance Analysis of Artificial Noise Assisted Secure Transmission for MISO systems with Multiple Eavesdroppers

学校读者我要写书评

暂无评论

Secrecy Performance Analysis of Artificial Noise Assisted Se...

2023 IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2023

作者： Xu, Panpan Tan, Weijie Niu, Kun Li, Zhen Peng, Changgen School of Mathematics and Statistics Guizhou University Guiyang China Key Laboratory of Advanced Manufacturing Technology of Ministry of Education Guizhou China Guizhou Big Data Academy Guizhou University Guiyang China College of Computer Science and Technology Guizhou University State Key Laboratory of Public Big Data Guiyang China College of Big Data and Information Engineering Guizhou University Guiyang China

ISBN: (纸本)9798350316728

In the paper, we investigate the secure communication of multiple-input single-output (MISO) systems with multiple eavesdroppers. We jointly design the beamforming (BF) and the artificial noise (AN) in MISO systems with multiple eavesdroppers to enhance the secrecy rate. Furthermore, under the secrecy interrupt constraint, we obtain the calculation formulas of the secrecy interrupt probability and the optimal secrecy rate. The simulation results demonstrate that AN with a higher dimension can effectively obstruct eavesdroppers and enhance the system's optimal secrecy rate. In addition, it is proved that the 1/4 full-dimension AN's secrecy rate close to that of the full-dimension AN in the system with multiple eavesdroppers, allowing the trade-off between system secrecy performance and artificial noise design complexity to be realized. © 2023 IEEE.

关键词： Economic and social effects

Complex-Valued Neural Network based Federated Learning for Multi-user Indoor Positioning Performance Optimization

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Yu, Hanzhi Chen, Mingzhe Liu, Yuchen The Department of Electrical and Computer Engineering Frost Institute for Data Science and Computing University of Miami Coral GablesFL33146 United States The Department of Computer Science North Carolina State University RaleighNC27695 United States

In this article, the use of channel state information (CSI) for indoor positioning is studied. In the considered model, a server equipped with several antennas sends pilot signals to users, while each user uses the received pilot signals to estimate channel states for user positioning. To this end, we formulate the positioning problem as an optimization problem aiming to minimize the gap between the estimated positions and the ground truth positions of users. To solve this problem, we design a complex-valued neural network (CVNN) model based federated learning (FL) algorithm. Compared to standard real-valued centralized machine learning (ML) methods, our proposed algorithm has two main advantages. First, our proposed algorithm can directly process complex-valued CSI data without data transformation. Second, our proposed algorithm is a distributed ML method that does not require users to send their CSI data to the server. Since the output of our proposed algorithm is complex-valued which consists of the real and imaginary parts, we study the use of the CVNN to implement two learning tasks. First, the proposed algorithm directly outputs the estimated positions of a user. Here, the real and imaginary parts of an output neuron represent the 2D coordinates of the user. Second, the proposed method can output two CSI features (i.e., line-of-sight/non-line-of-sight transmission link classification and time of arrival (TOA) prediction) which can be used in traditional positioning algorithms. Simulation results demonstrate that our designed CVNN based FL can reduce the mean positioning error between the estimated position and the actual position by up to 36%, compared to a RVNN based FL which requires to transform CSI data into real-valued data. Copyright © 2024, The Authors. All rights reserved.

关键词： Channel state information

Identifying Trustworthiness Challenges in Deep Learning Models for Continental-Scale Water Quality Prediction

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Xia, Xiaobo Liu, Xiaofeng Liu, Jiale Fang, Kuai Lu, Lu Oymak, Samet Currie, William S. Liu, Tongliang School of Computer Science University of Sydney Australia Department of Statistics and Data Science Yale University United States Michigan Institute for Data and AI in Society University of Michigan United States School for Environment and Sustainability University of Michigan United States College of Information Science and Technology Penn State University United States Department of Earth System Science Stanford University United States Department of Electrical Engineering and Computer Science University of Michigan United States

Water quality is foundational to environmental sustainability, ecosystem resilience, and public health. Deep learning models, particularly Long Short-Term Memory (LSTM) networks, offer trans-formative potential for large-scale water quality prediction and scientific insights generation. However, their widespread adoption in high-stakes decision-making, such as pollution mitigation and equitable resource allocation, is prevented by unresolved trustworthiness challenges including fairness, uncertainty, interpretability, robustness, generalizability, and reproducibility. In this work, we present the first comprehensive evaluation of trustworthiness in a continental-scale multi-task LSTM model predicting 20 water quality variables (encompassing physical/chemical processes, geochemical weathering, and nutrient cycling) across 482 U.S. basins. Our investigation uncovers systematic patterns of model performance disparities linked to basin characteristics, the inherent complexity of biogeochemical processes, and variable predictability, emphasizing critical performance fairness concerns. We further propose methodological frameworks for quantitatively evaluating critical aspects of trustworthiness, including uncertainty, interpretability, and robustness, identifying key limitations that could challenge reliable real-world deployment. This work serves as a timely call to action for advancing trustworthy data-driven methods for water resources management and provides a pathway to offering critical insights for researchers, decision-makers, and practitioners seeking to leverage artificial intelligence (AI) responsibly in environmental management. Copyright © 2025, The Authors. All rights reserved.

关键词： Resource allocation

Absolute Face Recognition System using Machine Learning Approach from Blurred Images

学校读者我要写书评

暂无评论

Absolute Face Recognition System using Machine Learning Appr...

Advances in Computing, Communication and Applied Informatics (ACCAI), International Conference on

作者： D. Hemavathi Utkarsh Yashwant Tambe Pao-Ann Hsiung Department of Data Science and Business Systems SRM Institute of Science and Technology Kattankulathur (Chennai) India Department of Computer Science and Information Engineering National Chung Cheng University Chiayi Taiwan R. O. C.

ISBN: (数字)9798350389449

ISBN: (纸本)9798350389456

Face Recognition is an important task in many domains to obtain the exact image from the pool of images. In general, blurred images are extremely challenging for the sensitive areas like law and order, defense, etc. Many face recognition techniques work well in normal images with various dimensions. In sensitive domains, the blurred image act as key evidence for the entire scenario. So, it is very important to find the original and accurate image from the blurred images. Representation of face, extraction of features and classification are the important steps in face recognition process. In existing models, Linear Binary Pattern (LBP) methods are used to recognize accurate images. LBP with Histogram (LBPH) is used to improve the detection performance of original images. Image sorting is done using the Point Spread Function Estimation on the blurred region and it helped to recognize faces with more accuracy. Extended Uniform Linear Binary Pattern method is used to reduce the dimensions to concentrate more on center pixels, with the use of the Viola-Jones algorithm and K-nearest neighbor (KNN) classifier for *** proposed enhanced LBP approach assisted in achieving 94.7% accuracy in recognizing human faces from blurry images.

关键词： Image recognition Accuracy Machine learning algorithms Face recognition Nearest neighbor methods Prediction algorithms Feature extraction

A Unified Framework for Convergence Analysis in Social Networks

学校读者我要写书评

暂无评论

A Unified Framework for Convergence Analysis in Social Netwo...

IEEE Conference on Decision and Control

作者： Ti-Chung Lee Jun-Kai Huang Youfeng Su Department of Electrical Engineering National Sun Yat-sen University Kaohsiung Taiwan College of Computer and Data Science Fuzhou University Fuzhou China

ISBN: (数字)9798350316339

ISBN: (纸本)9798350316346

This paper proposes a unified framework for the stability analysis of discrete-time nonlinear systems from social networks, including the Friedkin-Johnsen opinion model, two opinion dynamics models in the study of social power, and a general nonlinear polar opinion model. Three novel convergence results are proposed to treat various conditions based on LaSalle invariance principle. Several applications are provided to illustrate the power of the proposed framework.

关键词： Analytical models Social networking (online) Stability analysis Nonlinear systems Convergence

Temporal Gated Face Alignment Network for Camera-Based Physiological Sensing

学校读者我要写书评

暂无评论

SSRN

SSRN 2024年

作者： Li, Qi Guo, Dan Qian, Wei Zhou, Yuanen Zhao, Haifeng Sun, Xiao Wang, Meng School of Computer Science and Technology Anhui University Hefei230601 China Key Laboratory of Knowledge Engineering with Big Data Ministry of Education School of Computer Science and Information Engineering Hefei University of Technology Hefei230601 China Institute of Artificial Intelligence Hefei Comprehensive National Science Center Hefei230026 China

The remote photoplethysmography (rPPG) technique enables the estimation of vital signs such as heart rate by analyzing pulse-induced subtle skin color variation from facial videos. Robustly deriving cardiac pulse information with head motion (\eg, rotation and swings) interference, remains a critical challenge, as it inevitably introduces complex noise disturbances such as facial geometric deformation and head motions. Most existing approaches primarily focus on how to extract the weak pulse signals, overlooking the detrimental effects induced by such noises. In response to this, we propose a Temporal Gated Face Alignment Network (TGFAN) to adaptively counteract motion noise in long-term video sequences. As key modules in TGFAN, the bidirectional temporal face alignment block first captures inter-frame motion discrepancies and achieves temporal alignment, and then extracts motion-robust pulse features by spatiotemporal convolution. Furthermore, we also propose a temporal gating mechanism, which can dynamically guide motion-disturbed segments for face alignment to alleviate local temporal interference. Experimental evaluations on four public datasets demonstrate the superior performance of our proposed model under both intra-dataset and cross-dataset tests. Our codes will be available at \url{https://***/Eadith/TGFAN}. © 2024, The Authors. All rights reserved.

关键词： Photoplethysmography

UAV-assisted Joint Mobile Edge Computing and data Collection via Matching-enabled Deep Reinforcement Learning

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Wang, Boxiong Kang, Hui Li, Jiahui Sun, Geng Sun, Zemin Wang, Jiacheng Niyato, Dusit College of Computer Science and Technology Jilin University Changchun130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun130012 China College of Computing and Data Science Nanyang Technological University 639798 Singapore College of Computing and Data Science Nanyang Technological University Singapore

Unmanned aerial vehicle (UAV)-assisted mobile edge computing (MEC) and data collection (DC) have been popular research issues. Different from existing works that consider MEC and DC scenarios separately, this paper investigates a multi-UAV-assisted joint MEC-DC system. Specifically, we formulate a joint optimization problem to minimize the MEC latency and maximize the collected data volume. This problem can be classified as a non-convex mixed integer programming problem that exhibits long-term optimization and dynamics. Thus, we propose a deep reinforcement learning-based approach that jointly optimizes the UAV movement, user transmit power, and user association in real time to solve the problem efficiently. Specifically, we reformulate the optimization problem into an action space-reduced Markov decision process (MDP) and optimize the user association by using a two-phase matching-based association (TMA) strategy. Subsequently, we propose a soft actor-critic (SAC)-based approach that integrates the proposed TMA strategy (SAC-TMA) to solve the formulated joint optimization problem collaboratively. Simulation results demonstrate that the proposed SAC-TMA is able to coordinate the two subsystems and can effectively reduce the system latency and improve the data collection volume compared with other benchmark algorithms. Copyright © 2025, The Authors. All rights reserved.

关键词： Unmanned aerial vehicles (UAV)

An Online Chronic Disease Prediction System Based on Incremental Deep Neural Network

学校读者我要写书评

暂无评论

computers, Materials & Continua 2021年第4期67卷 951-964页

作者： Bin Yang Lingyun Xiang Xianyi Chen Wenjing Jia School of Design Jiangnan UniversityWuxi214122China School of Computer and Communication Engineering Changsha University of Science and TechnologyChangsha410114China Jiangsu Engineering Center of Network Monitoring Nanjing University of Information Science and TechnologyNanjing210044China School of Electrical and Data Engineering University of Technology SydneySydney2000Australia

Many chronic disease prediction methods have been proposed to predict or evaluate diabetes through artificial neural ***,due to the complexity of the human body,there are still many challenges to face in that *** of them is how to make the neural network prediction model continuously adapt and learn disease data of different patients,*** paper presents a novel chronic disease prediction system based on an incremental deep neural *** propensity of users suffering from chronic diseases can continuously be evaluated in an incremental *** time,the system can predict diabetes more and more accurately by processing the feedback *** diabetes prediction studies are based on a common dataset,the Pima Indians diabetes dataset,which has only eight input *** order to determine the correlation between the pathological characteristics of diabetic patients and their daily living resources,we have established an in-depth cooperation with a hospital.A Chinese diabetes dataset with 575 diabetics was ***’data collected by different sensors were used to train the network *** evaluated our system using a real-world diabetes dataset to confirm its *** experimental results show that the proposed system can not only continuously monitor the users,but also give early warning of physiological data that may indicate future diabetic ailments.

关键词： Deep learning incremental learning network architecture design chronic disease prediction