检索结果-内蒙古大学图书馆

32nd ACM International Conference on Multimedia, MM 2024

作者： Wu, Song Wei, Xiaoyu Chen, Xinyue Ren, Yazhou He, Jing Pu, Xiaorong School of Computer Science and Engineering University of Electronic Science and Technology of China Sichuan Chengdu China Shenzhen Institute for Advanced Study University of Electronic Science and Technology of China Guangdong Shenzhen China Faculty of Medicine & Biomedical Sciences The University of Queensland BrisbaneQLD Australia

ISBN: (纸本)9798400706868

Semi-supervised medical image segmentation has gained increasing attention due to its potential to alleviate the manual annotation burden. Mainstream methods typically involve two subnets, and conduct a consistency objective to ensure them producing consistent predictions for unlabeled data. However, they often ignore that the complementarity of model predictions is equally crucial. To realize the potential of the multi-subnet architecture, we propose a novel cross-view mutual learning method with a two-branch co-training framework. Specifically, we first introduce a novel conflict-based feature learning (CFL) that encourages the two subnets to learn distinct features from the same input. These distinct features are then decoded into complementary model predictions, allowing both subnets to understand the input from different views. More importantly, we propose a cross-view mutual learning (CML) to maximize the effectiveness of CFL. This approach requires only modifications to the model inputs and supervisory signals, and implements a heterogeneous consistency objective to fully explore the complementarity of model predictions. Consequently, the aggregated predictions can effectively capture both consistency and complementarity across two subnets. Experimental results on three public datasets demonstrate the superiority of CML over previous SoTA methods. Code is available at https://***/SongwuJob/CML. © 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

SensoryT5: Infusing Sensorimotor Norms into T5 for Enhanced Fine-grained Emotion Classification 8

SensoryT5: Infusing Sensorimotor Norms into T5 for Enhanced ...

引用

8th Workshop on Cognitive Aspects of the Lexicon, CogALex 2024

作者： Xia, Yuhan Zhao, Qingqing Long, Yunfei Xu, Ge Wang, Jia School of Computer Science and Electronic Engineering University of Essex United Kingdom Institute of Linguistics Chinese Academy of Social Sciences China College of Computer and Control Engineering Minjiang University China Department of Intelligent Science Xi’an Jiaotong-Liverpool University China

ISBN: (纸本)9782493814456

In traditional research approaches, sensory perception and emotion classification have traditionally been considered separate domains. Yet, the significant influence of sensory experiences on emotional responses is undeniable. The natural language processing (NLP) community has often missed the opportunity to merge sensory knowledge with emotion classification. To address this gap, we propose SensoryT5, a neuro-cognitive approach that integrates sensory information into the T5 (Text-to-Text Transfer Transformer) model, designed specifically for fine-grained emotion classification. This methodology incorporates sensory cues into the T5’s attention mechanism, enabling a harmonious balance between contextual understanding and sensory awareness. The resulting model amplifies the richness of emotional representations. In rigorous tests across various detailed emotion classification datasets, SensoryT5 showcases improved performance, surpassing both the foundational T5 model and current state-of-the-art works. Notably, SensoryT5’s success signifies a pivotal change in the NLP domain, highlighting the potential influence of neuro-cognitive data in refining machine learning models’ emotional sensitivity. © 2024 ELRA Language Resource Association.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Finite-time robust performance improvement for non-linear systems in the presence of input non-linearity and external disturbance

引用

IET Control Theory and Applications 2023年第6期17卷 672-682页

作者： Wang, Haoping Ghaffari, Valiollah Mobayen, Saleh Bartoszewicz, Andrzej Fekih, Afef Assawinchaichote, Wudhichai School of Automation Nanjing University of Science and Technology Nanjing China Department of Electrical Engineering Persian Gulf University Bushehr Iran Graduate School of Intelligent Data Science National Yunlin University of Science and Technology Yunlin Douliou Taiwan Institute of Automatic Control Lodz University of Technology Lodz Poland Department of Electrical and Computer Engineering University of Louisiana at Lafayette LafayetteLA United States Department of Electronic and Telecommunication Engineering Faculty of Engineering King Mongkut's University of Technology Thonburi Bangkok Thailand

The main focus of this research is on improving the performance of dynamic systems with actuator non-linearities and time-varying disturbances. To this end, using the concept of finite-time stability, a novel observer is presented to estimate the disturbances in systems with input non-linearities. Then, introducing an innovative sliding manifold, a robust observer-based technique is derived to guarantee the finite-time stability of the sliding surface. Hence, the closed-loop system would be stable in the presence of hard non-linearities and time-varying disturbances. The performance of the proposed approach was assessed in using a simulation study of two numerical examples. The obtained results and comparison with a conventional method confirmed the benefits of the suggested approach. © 2022 The Authors. IET Control Theory & Applications published by John Wiley & Sons Ltd on behalf of The Institution of engineering and Technology.

关键词： Closed loop systems

来源：评论

学校读者我要写书评

暂无评论

Spiking Transformer with Experts Mixture 38

Spiking Transformer with Experts Mixture

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Zhou, Zhaokun Lu, Yijie Jia, Yanhao Che, Kaiwei Niu, Jun Huang, Liwei Shi, Xinyu Zhu, Yuesheng Li, Guoqi Yu, Zhaofei Yuan, Li School of Electronic and Computer Engineering Shenzhen Graduate School Peking University China Peng Cheng Laboratory China College of Computing and Data Science Nanyang Technological University Singapore School of Computer Science Peking University China Institute for Artificial Intelligence Peking University China Institute of Automation Chinese Academy of Sciences China Deep NeuroCognition Lab I2R and CFAR Agency for Science Technology and Research Singapore

Spiking Neural Networks (SNNs) provide a sparse spike-driven mechanism which is believed to be critical for energy-efficient deep learning. Mixture-of-Experts (MoE), on the other side, aligns with the brain mechanism of distributed and sparse processing, resulting in an efficient way of enhancing model capacity and conditional computation. In this work, we consider how to incorporate SNNs' spike-driven and MoE's conditional computation into a unified framework. However, MoE uses softmax to get the dense conditional weights for each expert and TopK to hard-sparsify the network, which does not fit the properties of SNNs. To address this issue, we reformulate MoE in SNNs and introduce the Spiking Experts Mixture Mechanism (SEMM) from the perspective of sparse spiking activation. Both the experts and the router output spiking sequences, and their element-wise operation makes SEMM computation spike-driven and dynamic sparse-conditional. By developing SEMM into Spiking Transformer, the Experts Mixture Spiking Attention (EMSA) and the Experts Mixture Spiking Perceptron (EMSP) are proposed, which performs routing allocation for head-wise and channel-wise spiking experts, respectively. Experiments show that SEMM realizes sparse conditional computation and obtains a stable improvement on neuromorphic and static datasets with approximate computational overhead based on the Spiking Transformer baselines. © 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

Semi-FLChain: Toward Secure and Efficient Semi-Supervised Federated Learning for IoT Using Blockchain 5

Semi-FLChain: Toward Secure and Efficient Semi-Supervised Fe...

引用

5th IEEE International Conference on Power, Intelligent Computing and Systems, ICPICS 2023

作者： Qi, Yuanhang Liu, Guisong School of Computer Science and Engineering University of Electronic Science and Technology of China Sichuan Chengdu611731 China School of Computer Science University of Electronic Science and Technology of China Zhongshan Institute Guangdong Zhongshan528402 China School of Computing and Artificial Intelligence Southwestern University of Finance and Economics Sichuan Chengdu611130 China

ISBN: (纸本)9798350333442

This paper proposes a novel semi-supervised federated learning (SSFL) algorithm for Internet of Things (IoT) systems using blockchain technology, called Semi-FLChain. The algorithm is designed to improve the efficiency and security of SSFL by combining hybrid data augmentation and Bayesian estimation to enhance the consistency regularization and pseudo-labeling methods in SSFL. The proposed algorithm aims to overcome the limitations of existing SSFL methods, such as the low accuracy of pseudo-labeling and the high computational cost of consistency regularization. The use of blockchain technology in Semi-FLChain further ensures the security and privacy of IoT devices and their data during the training process. The experimental results demonstrate that Semi-FLChain outperforms state-of-the-art SSFL algorithms in terms of both accuracy and efficiency, making it a promising solution for secure and efficient federated learning in IoT systems. © 2023 IEEE.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

Optimal Deployment of Heterogeneous Visual Sensor Networks for Target Coverage and Feature Aggregation with Lifetime Guarantee 21

Optimal Deployment of Heterogeneous Visual Sensor Networks f...

引用

21st International Conference on Networking, Sensing and Control, ICNSC 2024

作者： Zhu, Xiaojian Zhou, MengChu School of Computer Science and Engineering Changshu Institute of Technology Changshu215500 China School of Information and Electronic Engineering Zhejiang Gongshang University Hangzhou310018 China New Jersey Institute of Technology Department of Electrical and Computer Engineering NewarkNJ07102 United States

ISBN: (纸本)9798350365221

Due to their need to handle a huge size of image data and limited battery power at each node, visual sensor networks (VSNs) suffer from short network lifetime and long network delay. Consequently, how to reduce network traffic is very important in their operation. The deployment of processing nodes for feature extraction is an effective way to reduce it. Deploying multiple types of camera nodes should help do so as well. Therefore, in this work we investigate how to deploy a heterogeneous VSN for target coverage and feature aggregation with lifetime guarantee at the minimum cost. Since this problem is NP-hard, we propose two heuristics, i.e., separate and joint node deployment algorithms, to solve it. A large number of experimental results expose that the latter has higher success rate and can obtain lower network deployment cost than the former at the expense of extra computation. © 2024 IEEE.

关键词： Sensor nodes

来源：评论

学校读者我要写书评

暂无评论

HGNNDroid: Android Malware Detection Based on Heterogeneous Graph Neural Network 9

HGNNDroid: Android Malware Detection Based on Heterogeneous ...

引用

9th IEEE International Conference on data science in Cyberspace, DSC 2024

作者： Liu, Xingyu Liu, Xiaozhen Hao, Kegang Wang, Ke Chen, Xinglong Niu, Weina University of Electronic Science and Technology of China School of Computer Science and Engineering Chengdu China Chengdu Wonder-Explorer Technology Co. Ltd. Chengdu China Southwest China Research Institute of Electronic Equipment Chengdu China

ISBN: (纸本)9798350391367

With the rapid development of mobile internet, smartphones have become indispensable tools in people's daily lives. As the predominant mobile operating system, Android records various user information, making it a prime target for malicious attacks. In response to the increasing threat of malware, scholars have proposed various feature extraction methods and innovative models to detect Android malware. Researchers have introduced detection methods based on heterogeneous graph to leverage the intrinsic information and the correlation information of samples. However, existing methods lack sufficient mining of semantic information from feature nodes. To address this issue, we propose HGNNDroid, an Android malware detection method based on hybrid analysis and a heterogeneous graph neural network. It utilizes hybrid analysis to extract features from APK files, constructing a more comprehensive behavioral profile of the samples. Subsequently, a heterogeneous graph neural network with a two-layer semantic fusion mechanism performs the classification task. The network aggregates the semantic information of features from different perspectives and utilizes meta-paths to discover the potential structural relationships, thereby achieving more accurate Android malware detection. On the CICMalDroid 2020 dataset, the F1-score reached 98.72%, outperforming HAWK and five other methods. © 2024 IEEE.

关键词： Android malware

来源：评论

学校读者我要写书评

暂无评论

Balancing Usability and Security Innovative Approaches to Privacy in Digital Environments

Balancing Usability and Security Innovative Approaches to Pr...

引用

2024 International Conference on Communication, computer sciences and engineering, IC3SE 2024

作者： Baswaraju, Swathi Yadav, M. Geeta Nijhawan, Ginni Rana, Ajay Rani, T. K. Pushpa Al-Rubaye, Taqi Mohammed Khattab New Horizon College Of Engineering Department Of Computer Science - Data Science Bangalore India Institute Of Aeronautical Engineering Department Of Computer Science And Engineering Telangana Hyderabad India Lovely Professional University Phagwara India Amity School Of Engineering And Technology Amity University Greater Noida India Mlr Institute Of Technology Department Of Computer Science And Engineering Telangana Hyderabad India College Of Medical Technologies The Islamic University Department Of Medical Laboratory Technology Najaf Iraq

ISBN: (纸本)9798350366846

data analysis, private protection, detecting unusual things, identifying people, and transferring data are part of the study's unique security design. The major system method is AUA. Your authentication level depends on your activity, device, and location. Context-aware Privacy uses AUA to tailor user privacy options. Encrypting AES data transfers with Secure data Transfer (SDT) is offered. Machine learning anomaly detection (ADML) analyzes user behavior and implements security measures to uncover vulnerabilities rapidly. DPDA employs controlled noise to make it statistically tougher to identify a person while processing data to safeguard privacy. Ablation study indicates adaptive security design strategies function better together. The recommended design may be safer and easier to employ with visual models and performance testing. With this knowledge, managers may be able to create business-friendly security solutions. Finally, the framework's extensive and adaptable security models boost digital safety. It offers new privacy protection in the internet age. © 2024 IEEE.

关键词： data analytics

来源：评论

学校读者我要写书评

暂无评论

Recent Challenges in Medical science using Machine Learning Techniques: A Review 7

Recent Challenges in Medical Science using Machine Learning ...

引用

7th International Conference on I-SMAC (IoT in Social, Mobile, analytics and Cloud), I-SMAC 2023

作者： Shah, Milind Padhiyar, Priyanka Parmar, Kajal Patel, Jemisha Panesar, Sonia Raja, Krupa Department of Computer Engineering Changa388421 India Department of Computer Science & Engineering Changa388421 India Vadodara Institute of Engineering Department of Computer Engineering Gujarat Vadodara India Krishna School of Emerging Department of Computer Science & Engineering Gujarat Vadodara India Engineer Decimal Point Analytics Gujarat Vadodara India

ISBN: (纸本)9798350341485

Medical science is significantly impacted by the applications of Machine Learning (ML). ML is a subset of Artificial Intelligence (AI) technology that improves physicians' efficiency and precision. Recently, machine learning techniques have facilitated a tremendous expansion in the field of medicine. The field of medical science has been revolutionized by machine learning, which provides new methods for analyzing large-scale patient data and providing accurate predictions for diagnosis, treatment, and disease progression. The risk of developing diseases such as cancer, diabetes, and cardiovascular disease has been predicted using machine learning algorithms. These algorithms analyze large sets of patient data, including genetic information, medical history, and lifestyle factors, to recognize patterns and predict disease risk. In addition, machine learning has been used to develop individualized treatment programs for patients. By analyzing patient data, machine learning algorithms can determine the most effective treatments for each patient, thereby reducing the risk of adverse effects and enhancing patient outcomes. Moreover, machine learning techniques have been used to develop image analysis tools for medical imaging, allowing doctors to accurately diagnose diseases such as cancer and identify the location and severity of injuries. This review discusses the current challenges of medical science using machine learning techniques and how medical science problems can be solved using machine learning algorithms. © 2023 IEEE.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Detecting UAV Target with Deep Convolutional Neural Network 24

Detecting UAV Target with Deep Convolutional Neural Network

引用

6th International electronics Communication Conference, IECC 2024

作者： Norbelt, Mutagisha Luo, Xiling Sun, Jinping Wang, Junjun Claude, Uwimana Wisal, Muhammad Hangzhou Innovation Institute Beihang University Hangzhou China School of Electronic and Information Engineering Beihang University Beijing China School of Computer Science Beihang University Beijing China

ISBN: (纸本)9798400717598

Over the past years, Convolutional Neural Networks (CNNs) have emerged as widely used techniques for tackling various computer vision challenges, with image detection being a popular application. The performance of CNNs in object detection has rapidly improved in terms of both accuracy and speed. Unmanned Aerial Vehicles (UAVs), or drones, have seen exponential growth in adoption across various sectors, from military and surveillance to agriculture and infrastructure inspection. Their ability to capture aerial data and perform remote sensing tasks has revolutionized industries, making them invaluable tools for monitoring, mapping, and data collection. However, UAVs flying near runways can pose a danger to airplanes during take-off and landing. Additionally, drones can be a threat to stadiums, prisons, and military camps due to their ability to carry payloads and bypass ground security. This research concentrates on a particular aspect of object detection: identifying unmanned aerial vehicles (UAVs or drones). The task becomes more demanding due to the data being extracted from video footage captured simultaneously by a visible light camera and an infrared camera. This setup results in difficult viewing angles and significant motion blur. To assess both accuracy and speed, the study selected two models with typical architectures: You Only Look Once (YOLO) and Single Shot MultiBox Detector (SSD). The goal was to evaluate the robustness of these models in detecting UAVs under conditions of motion blur and to understand how the structural differences between the models might influence the results. Several experiments were conducted as part of this work to achieve the stated objectives. The findings likely provide valuable insights into the performance of these object detection models when applied to the challenging scenario of UAV detection from video streams with motion blur. The results showed that for better structural design, YOLOv3 outperformed SSD in various aspec

关键词： Drones

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：