检索结果-内蒙古大学图书馆

A recover-then-discriminate framework for robust anomaly detection

science China(Information sciences) 2025年第4期68卷 300-318页

作者： Peng XING Dong ZHANG Jinhui TANG Zechao LI School of Computer Science and Engineering Nanjing University of Science and Technology Department of Electronic and Computer Engineering The Hong Kong University of Science and Technology

Anomaly detection(AD) has been extensively studied and applied across various scenarios in recent years. However, gaps remain between the current performance and the desired recognition accuracy required for practical *** paper analyzes two fundamental failure cases in the baseline AD model and identifies key reasons that limit the recognition accuracy of existing approaches. Specifically, by Case-1, we found that the main reason detrimental to current AD methods is that the inputs to the recovery model contain a large number of detailed features to be recovered, which leads to the normal/abnormal area has not/has been recovered into its original state. By Case-2, we surprisingly found that the abnormal area that cannot be recognized in image-level representations can be easily recognized in the feature-level representation. Based on the above observations, we propose a novel recover-then-discriminate(ReDi) framework for *** takes a self-generated feature map(e.g., histogram of oriented gradients) and a selected prompted image as explicit input information to address the identified in Case-1. Additionally, a feature-level discriminative network is introduced to amplify abnormal differences between the recovered and input representations. Extensive experiments on two widely used yet challenging AD datasets demonstrate that ReDi achieves state-of-the-art recognition accuracy.

关键词： recovery network HOG prompt discriminative network self-correlation loss anomaly detection

来源：评论

学校读者我要写书评

暂无评论

Data-driven output regulation control for constrained linear systems

引用

science China(Information sciences) 2025年第3期68卷 338-353页

作者： Chaoyu XIA Yi DONG Chaoli WANG Shengyuan XU Shanghai Research Institute for Intelligent Autonomous Systems Tongji University College of Electronic and Information Engineering Shanghai Research Institute for Intelligent Autonomous SystemsTongji University Department of Control Science and Engineering School of Optical-Electrical and Computer EngineeringUniversity of Shanghai for Science and Technology Department of Automation Nanjing University of Science and Technology

This study introduces a data-driven approach for state and output feedback control addressing the constrained output regulation problem in unknown linear discrete-time systems. Our method ensures effective tracking performance while satisfying the state and input constraints, even when system matrices are not available. We first establish a sufficient condition necessary for the existence of a solution pair to the regulator equation and propose a data-based approach to obtain the feedforward and feedback control gains for state feedback control using linear programming. Furthermore, we design a refined Luenberger observer to accurately estimate the system state, while keeping the estimation error within a predefined set. By combining output regulation theory, we develop an output feedback control strategy. The stability of the closed-loop system is rigorously proved to be asymptotically stable by further leveraging the concept of λ-contractive sets.

关键词： output regulation constrained system data-driven.

来源：评论

学校读者我要写书评

暂无评论

Enhanced Acceleration for Generalized Nonconvex Low-Rank Matrix Learning

引用

Chinese Journal of electronics 2025年第1期34卷 98-113页

作者： Hengmin Zhang Jian Yang Wenli Du Bob Zhang Zhiyuan Zha Bihan Wen School of Electrical and Electronic Engineering Nanyang Technological University School of Computer Science and Engineering Nanjing University of Science and Technology School of Information Science and Engineering East China University of Science and Technology Department of Electrical and Computer Engineering University of Macau

Matrix minimization techniques that employ the nuclear norm have gained recognition for their applicability in tasks like image inpainting, clustering, classification, and reconstruction. However, they come with inherent biases and computational burdens, especially when used to relax the rank function, making them less effective and efficient in real-world scenarios. To address these challenges, our research focuses on generalized nonconvex rank regularization problems in robust matrix completion, low-rank representation, and robust matrix regression. We introduce innovative approaches for effective and efficient low-rank matrix learning, grounded in generalized nonconvex rank relaxations inspired by various substitutes for the ?0-norm relaxed functions. These relaxations allow us to more accurately capture low-rank structures. Our optimization strategy employs a nonconvex and multi-variable alternating direction method of multipliers, backed by rigorous theoretical analysis for complexity and *** algorithm iteratively updates blocks of variables, ensuring efficient convergence. Additionally, we incorporate the randomized singular value decomposition technique and/or other acceleration strategies to enhance the computational efficiency of our approach, particularly for large-scale constrained minimization problems. In conclusion, our experimental results across a variety of image vision-related application tasks unequivocally demonstrate the superiority of our proposed methodologies in terms of both efficacy and efficiency when compared to most other related learning methods.

关键词： Learning systems Image recognition Minimization Computational efficiency Complexity theory Matrix decomposition Optimization Image reconstruction Singular value decomposition Convergence

来源：评论

学校读者我要写书评

暂无评论

Modeling Task Engagement to Regulate Reinforcement Learning-based Decoding for Online Brain Control

引用

IEEE Transactions on Cognitive and Developmental systems 2024年第3期17卷 606-614页

作者： Zhang, Xiang Shen, Xiang Wang, Yiwen Department of Electronic and Computer Engineering Hong Kong University of Science and Technology Hong Kong Department of Chemical and Biological Engineering Department of Electronic and Computer Engineering Hong Kong University of Science and Technology Hong Kong

Brain-Machine Interfaces (BMIs) offer significant promise for enabling paralyzed individuals to control external devices using their brain signals. One challenge is that during the online Brain Control (BC) process, subjects may not be completely immersed in the task, particularly when multiple steps are needed to achieve a goal. The decoder indiscriminately takes the less engaged trials as training data, which might decrease the decoding accuracy. In this paper, we propose an alternative kernel RL-based decoder that trains online with continuous parameter update. We model neural activity from the medial prefrontal cortex (mPFC), a reward-related brain region, to represent task engagement. This information is incorporated into a stochastic learning rate using an exponential model, which measures the relevancy of neural data. The proposed algorithm was evaluated in the experiment where rats performed a cursor-reaching BC task. We found the neural activities from mPFC contained the engagement information which was negatively correlated with trial response time. Moreover, compared to the RL method without task engagement modeling, our proposed method enhanced the training efficiency. It used half of the training data to achieve the same reconstruction accuracy of the cursor trajectory. The results demonstrate the potential of our RL framework for improving online brain control tasks. © 2016 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Efficient 3D Tracking and Detection of Multi-Rotor UAVs Using mmWave Radar with Semi-Supervised Learning

引用

IEEE Sensors Journal 2025年第12期25卷 22001-22014页

作者： Xi, Rui Wei, Wenjie Zhang, Malu University of Electronic Science and Technology of China Department of Computer Science and Engineering Chengdu611731 China

Small UAVs pose security risks to sensitive areas and individuals due to their rapid movement and wide coverage capabilities. Effective monitoring necessitates the deployment of lightweight and energy-efficient surveillance systems. This research introduces an efficient 3D tracking and detection approach for small UAVs, utilizing millimeter-wave radars and Spiking Neural Networks (SNNs). By capturing micro-Doppler characteristics of UAV movements, it effectively processes low signal-to-noise ratios and uncertain signals. An improved angle estimation algorithm, combining dynamic programming and particle filters, enables realtime 3D UAV tracking with reduced computational complexity. Then, a simple UAV detection model based on SNN architecture is developed by leveraging UAVs’ position and corresponding Doppler information. Furthermore, a bio-inspired semi-supervised method is proposed to facilitate the training of SNNs using a limited number of annotated samples. The effectiveness of the proposed methodology is evaluated under various environmental conditions. Results indicate a significant improvement in tracking computation time efficiency, with the recognition model size reduced to one-tenth of its original size, yet it maintains near-original system performance. © 2001-2012 IEEE.

关键词： Semi-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Privacy-preserving recommendation with coarse-grained spatiotemporal contexts

引用

science China(Information sciences) 2025年第4期68卷 66-81页

作者： Lei CHEN Chen GAO Jiahuan LEI Xiaoyi DU Xinlei SHI Hengliang LUO Depeng JIN Yong LI Meng WANG Department of Electronic Engineering BNRist Tsinghua University Meituan Inc. School of Computer Science and Information Engineering Hefei University of Technology

The behavior of users on online life service platforms like Meituan and Yelp often occurs within specific finegrained spatiotemporal contexts(i.e., when and where). Recommender systems, designed to serve millions of users, typically operate in a fully server-based manner, requiring on-device users to upload their behavioral data, including fine-grained spatiotemporal contexts, to the server, which has sparked public concern regarding privacy. Consequently, user devices only upload coarse-grained spatiotemporal contexts for user privacy protection. However, previous research mostly focuses on modeling fine-grained spatiotemporal contexts using knowledge graph convolutional models, which are not applicable to coarse-grained spatiotemporal contexts in privacy-constrained recommender systems. In this paper, we investigate privacy-preserving recommendation by leveraging coarse-grained spatiotemporal contexts. We propose the coarse-grained spatiotemporal knowledge graph for privacy-preserving recommendation(CSKG), which explicitly models spatiotemporal co-occurrences using common-sense knowledge from coarse-grained contexts. Specifically, we begin by constructing a spatiotemporal knowledge graph tailored to coarse-grained spatiotemporal contexts. Then we employ a learnable metagraph network that integrates common-sense information to filter and extract co-occurrences. CSKG evaluates the impact of coarsegrained spatiotemporal contexts on user behavior through the use of a knowledge graph convolutional network. Finally, we introduce joint learning to effectively learn representations. By conducting experiments on two real large-scale datasets,we achieve an average improvement of about 11.0% on two ranking metrics. The results clearly demonstrate that CSKG outperforms state-of-the-art baselines.

关键词： privacy-preserveing coarse-grained spatiotemporal contexts recommender systems

来源：评论

学校读者我要写书评

暂无评论

Underwater object detection based on enhanced YOLOv4 architecture

引用

Multimedia Tools and Applications 2024年第18期83卷 53759-53783页

作者： Liu, Ching-Hua Lin, Chang Hong Department of Electronic and Computer Engineering National Taiwan University of Science and Technology Taiwan

Object detection and image restoration pose significant challenges in deep learning and computer vision. These tasks are widely employed in various applications, and there is an increasing demand for specialized environments where images are prone to blur or noise, which can adversely affect subsequent results. In recent years, significant breakthroughs have been achieved in object detection performance. While some previously proposed methods prioritize high accuracy at the cost of longer inference times, others prioritize speed. Therefore, it is crucial to design an efficient network architecture that maintains both inference speed and high accuracy. This research proposes a network architecture for underwater object detection with an attention mechanism. The proposed approach differentiates itself from other methods by employing a deblurring network as a preprocessing step to restore and enhance the image quality of the underwater dataset. Additionally, in the feature extraction stage of the detection network, channel and spatial feature information are individually enhanced. These adaptive attention features are then integrated into a multi-scale feature fusion. Finally, the cross-stage local method is combined to improve the learning ability of the convolutional neural network while reducing the size of the model. Based on the experimental results, our proposed model architecture achieves leading accuracy and strikes a favorable balance in terms of model size compared to previously proposed methods. Based on our structure, the metrics of AP and AP50 reach 66.8 and 87.6, respectively. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2023.

关键词： Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

MMInstruct: a high-quality multi-modal instruction tuning dataset with extensive diversity

引用

science China(Information sciences) 2024年第12期67卷 36-51页

作者： Yangzhou LIU Yue CAO Zhangwei GAO Weiyun WANG Zhe CHEN Wenhai WANG Hao TIAN Lewei LU Xizhou ZHU Tong LU Yu QIAO Jifeng DAI School of Computer Science Nanjing University School of Electronic Information and Electrical Engineering Shanghai Jiao Tong University Shanghai AI Laboratory School of Computer Science Fudan University Department of Information Engineering The Chinese University of Hong Kong SenseTime Research Department of Electronic Engineering Tsinghua University

Despite the effectiveness of vision-language supervised fine-tuning in enhancing the performance of vision large language models(VLLMs), existing visual instruction tuning datasets include the following limitations.(1) Instruction annotation quality: despite existing VLLMs exhibiting strong performance,instructions generated by those advanced VLLMs may still suffer from inaccuracies, such as hallucinations.(2) Instructions and image diversity: the limited range of instruction types and the lack of diversity in image data may impact the model's ability to generate diversified and closer to real-world scenarios outputs. To address these challenges, we construct a high-quality, diverse visual instruction tuning dataset MMInstruct,which consists of 973k instructions from 24 domains. There are four instruction types: judgment, multiplechoice, long visual question answering, and short visual question answering. To construct MMInstruct, we propose an instruction generation data engine that leverages GPT-4V, GPT-3.5, and manual correction. Our instruction generation engine enables semi-automatic, low-cost, and multi-domain instruction generation at 1/6 the cost of manual construction. Through extensive experiment validation and ablation experiments,we demonstrate that MMInstruct could significantly improve the performance of VLLMs, e.g., the model fine-tuning on MMInstruct achieves new state-of-the-art performance on 10 out of 12 benchmarks. The code and data shall be available at https://***/yuecao0119/MMInstruct.

关键词： instruction tuning multi-modal multi-domain dataset vision large language model

来源：评论

学校读者我要写书评

暂无评论

Drug–target interactions prediction based on similarity graph features extraction and deep learning

引用

Neural Computing and Applications 2025年第6期37卷 4303-4322页

作者： Torkey, Hanaa El-Behery, Heba Attia, Abdel-Fattah El-Fishawy, Nawal Computer Science and Engineering Department Faculty of Electronic Engineering Menoufia University Menouf Egypt Department of Computer Science and Engineering Faculty of Engineering Kafrelsheikh University Kafrelsheikh Egypt Department of Computer Science College of Computer Engineering and Sciences Prince Sattam Bin Abdulaziz University Al-Kharj Saudi Arabia

Identifying drug–target interactions (DTIs) is a critical step in both drug repositioning. The labor-intensive, time-consuming, and costly nature of classic DTI laboratory studies makes it imperative to create efficient computer algorithms to forecast possible DTIs. However, current computational approaches that predict potential drug–target interactions (DTIs) suffer from some limitations, like finding the best similarity measures or negative samples, and thus require substantial performance improvement. This study proposes an integrated approach based on feature representation and deep learning to predict DTIs. We extract the relevant features of drugs and proteins from heterogeneous networks using graph mining techniques. The proposed approach constructs a heterogeneous graph from the known drug–protein interactions, protein–protein, and drug–drug similarities. Then applying two feature extraction techniques to extract the features, then utilizing these features in training a deep learning model to predict the potential DTIs. Also, a novel algorithm is proposed to find the negative samples based on the drug and protein similarity matrices. Four Benchmark datasets are used to evaluate the proposed approach. Our approach achieves the highest AUC (area under the ROC curve) across all datasets (0.98) with around 2% increases over the existing methods. Experimental results demonstrate that our proposed approach outperforms the baseline methods in predicting DTI, and our negative sample-identifying algorithm could be established as a competitive solution. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Arrhythmia Detection by Using Chaos Theory with Machine Learning Algorithms

引用

computers, Materials & Continua 2024年第6期79卷 3855-3875页

作者： Maie Aboghazalah Passent El-kafrawy Abdelmoty M.Ahmed Rasha Elnemr Belgacem Bouallegue Ayman El-sayed Math and Computer Science Department Faculty of ScienceMenoufia UniversityShebin El-komEgypt College of Engineering Computer Science DepartmentEffat UniversityJeddahKingdom of Saudi Arabia Department of Computer Engineering College of Computer ScienceKing Khalid UniversityAbha61421Saudi Arabia Computer Science and Engineering Department Faculty of Electronic EngineeringMenoufia UniversityShebin El-komEgypt Climate Change Information Center and Expert Systems Agriculture Research CenterGizaEgypt

Heart monitoring improves life ***(ECGs or EKGs)detect heart *** learning algorithms can create a few ECG diagnosis processing *** first method uses raw ECG and time-series *** second method classifies the ECG by patient *** third technique translates ECG impulses into Q waves,R waves and S waves(QRS)features using richer *** ECG signals vary naturally between humans and activities,we will combine the three feature selection methods to improve classification accuracy and *** using all three approaches have not been examined till *** researchers found that Machine Learning(ML)techniques can improve ECG *** study will compare popular machine learning techniques to evaluate ECG *** algorithms—Support Vector Machine(SVM),Decision Tree,Naive Bayes,and Neural Network—compare categorization *** plus prior knowledge has the highest accuracy(99%)of the four ML *** characteristics failed to identify signals without chaos *** 99.8%classification accuracy,the Decision Tree technique outperformed all previous experiments.

关键词： ECG extraction ECG leads time series prior knowledge and arrhythmia chaos theory QRS complex analysis machine learning ECG classification

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：