检索结果-内蒙古大学图书馆

Domain Adaptation with Sample Relation Reinforcement 1

24th International Conference on Web Information Systems Engineering, WISE 2023

作者： Teng, Shaohua Liu, Wenjie Guo, Ruixi Zhang, Wei Zheng, Zefeng Teng, Luyao Chen, Tongbao School of Computer Science and Technology Guangdong University of Technology Guangzhou510006 China School of Electrical Engineering Xi’an Jiaotong University Xi’an710049 China School of Information Engineering Guangzhou Panyu Polytechnic Guangzhou511483 China School of Advanced Manufacturing Guangdong University of Technology Guangzhou510006 China

ISBN: (数字)9789819972548

ISBN: (纸本)9789819972531

Domain adaptation transfers knowledge from the source domain into the target domain by minimizing the distribution discrepancy between two domains. Recently, existing approaches ignore the position differences between samples within the same class. However, there are still two problems that need to be resolved. Firstly, marginal samples and central samples may interact with each other, which leads to samples within the same class aggregate toward the margin (loss common knowledge) or toward the center (loss specific knowledge). Secondly, large distribution differences may result in degraded performance in classification. To solve these two problems, we propose a domain adaptation method called Sample Relation Reinforcement (SRR). To mitigate the interaction between marginal samples and central samples, marginal samples detection (MSD) and central samples detection (CSD) are proposed. Specifically, MSD propose a triple distances metric among the sample pairs and class centers to detect the marginal samples and learn specific knowledge. CSD utilizes the nearest and second-nearest centroid of samples to detect the central samples and learn common knowledge. In addition, dual centroid detection (DCD) is proposed to alleviate the negative effects of distribution differences, which utilizes the class centers of the source and target domains to classify the target samples. Finally, the experimental results on four datasets demonstrate that the proposed method outperforms other competing approaches significantly. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Reinforcement

来源：评论

学校读者我要写书评

暂无评论

Spatial distribution information extraction of winter wheat in henan Province based on multi-source sentinel remote sensing data

Spatial distribution information extraction of winter wheat ...

引用

2024 International Conference on Remote Sensing, Mapping, and Geographic Information Systems, RSMG 2024

作者： Zhang, Ziqian Lv, Junye Cui, Tengfei Zhou, Qiongbo Jiang, Chuande Liu, Zhiyang Si, Qiliang Henan University of Science and Technology Software School No. 263 Kaiyuan Avenue Luolong District Henan Province Luoyang City471000 China School of Surveying and Land Information Engineering Henan Polytechnic University No. 2001 Shiji Road Jiaozuo454000 China Boston University Metropolitan College BostonMA02215 United States University of Southern California Gould School of Law CA90089 United States School of Water Conservancy North China University of Water Resources and Electric Power Zhengzhou450046 China College of Science Weihai264209 China

ISBN: (纸本)9781510685826

Timely and accurately acquiring spatial distribution information for winter wheat is crucial for formulating agricultural policies. However, the information extraction for traditional spatial distribution for winter wheat relies primarily on manual field surveys. This method is labour-intensive and subjective and less accurate. With that remote sensing technology continues to develop, new avenues for extracting distribution information for winter wheat have emerged. Nevertheless, previous studies often did not fully leverage the advantages of remote sensing data from multiple sources, resulting in lower extraction accuracy. To address this, this study focuses on henan Province, where Sentinel-2 optical imagery and Sentinel-1 radar imagery. Then support vector machine and random forest algorithms were used to construct winter wheat extraction model. The results indicate that the extraction accuracy of Sentinel-2 optical imaging surpasses Sentinel-1 radar imaging, with an overall accuracy of 82.04%, the producer's accuracy for winter wheat reaches 86.25%, the accuracy of user is 85.19%, the Kappa coefficient is 0.770. Further integration within the Sentinel-2 optical imagery with Sentinel-1 radar imagery significantly enhances extraction's accuracy. The overall accuracy improves to 90.19%, the accuracy for producers increases to 96.25%, user's accuracy rises to 95.06%, and the Kappa coefficient increases to 0.874. Additionally, by comparing different machine learning algorithms, it is observed that the algorithm random forest consistently outperforms the algorithm support vector machine in extraction accuracy regardless of the dataset. In this way, this study provides valuable guidance for obtaining information on the spatial distribution of winter wheat in henan Province. © 2024 SPIE.

关键词： Radar imaging

来源：评论

学校读者我要写书评

暂无评论

When Zero-Trust Meets Federated Learning

When Zero-Trust Meets Federated Learning

引用

2024 IEEE Global Communications Conference, GLOBECOM 2024

作者： Zhang, Xinran Wang, Dan Zhu, Yifei Chen, Weilong Chang, Zheng Han, Zhu The Hong Kong Polytechnic University Department of Computing Hong Kong Shanghai Jiao Tong University UM-SJTU Joint Institute Shanghai China University of Electronic Science and Technology of China School of Computer Science and Engineering Chengdu China University of Jyväskylä Faculty of Information Technology Jyväskylä Finland University of Houston Department of Electrical and Computer Engineering HoustonTX United States Kyung Hee University Department of Computer Science and Engineering Seoul Korea Republic of

ISBN: (纸本)9798350351255

Nowadays, Federated Learning (FL) has emerged as a promising and critical machine learning scheme to protect data privacy and reduce communication overhead. As the scale and connectivity expand in the FL system, enhancing the model's robustness against security threats from malicious clients grows ever more critical. An effective defensive solution involves selecting benign clients appropriately, thereby mitigating the vulnerability of the FL system to malicious attacks. However, clients exhibit varying behaviors over time, which complicates the task of accurately modeling their future trustworthiness. Moreover, blindly trusting clients with high trust values poses risks, given the potential for severe losses from betrayal. To tackle these problems, we propose a zero-trust policy in FL aimed at establishing continuous trust in each client while maintaining skepticism towards potential betrayal attacks. Specifically, we develop a Dirichlet-based trust evaluation technique to enable a comprehensive selection of trustworthy participants. This technique leverages the posterior distribution to estimate clients' trust values from their evolving behavior records over time. Then, we anticipate potential betrayal from a selected client and formulate a min-max optimization problem to minimize the worst-case betrayal loss, thereby boosting the system's betrayalaware robustness. Next, we convert this problem into a convex optimization problem and utilize the interior point method for resolution. We conduct extensive simulations to validate the efficacy of our proposed zero-trust policy in accurately assessing trust and enhancing the model's robustness to betrayal. © 2024 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

A Federated Learning Scheme Based on Reputation Evaluation and Blockchain 29

A Federated Learning Scheme Based on Reputation Evaluation a...

引用

29th IEEE International Conference on Parallel and Distributed Systems, ICPADS 2023

作者： Tang, Siyu An, Jian Sun, Xiangyan Hu, Xiancheng Wang, Feifei Xi'an Jiaotong University School of Computer Science and Technology Xi'an China Henan Jiaoyuan Engineering Technology Group Co. Ltd Zhengzhou China Shaanxi Province Key Laboratory of Computer Network Xi'an Jiaotong University Xi'an710049 China

ISBN: (纸本)9798350330717

Federated learning is an emerging distributed machine learning paradigm, where multiple data sources collaborate to train a model without sharing the original data, safeguarding user data privacy. Nevertheless, users, whether intentionally or unintentionally, may contribute low-quality data that can impact model quality or infer user information from the model parameters, potentially leading to privacy breaches. Selecting appropriate participants for training is still a challenge needing resolution. This paper introduces a federated learning scheme called FREB, which leverages a reputation evaluation mechanism and blockchain technology. The foundation of this scheme is blockchain, and it incorporates reputation as an indicator of participant reliability. Reputation is computed using a multi-weight subjective logic model. Before the training process begins, participants are chosen based on their reputation. Furthermore, the paper presents a novel model aggregation approach. By injecting noise into the uploaded model parameters, the scheme ensures that potential attackers cannot glean sensitive data information from the parameters. To validate the efficacy of FREB, the authors conducted simulation experiments using real datasets. The results of these experiments demonstrate that FREB is capable of thwarting malicious node attacks and promoting high-quality collaborative learning. It achieves this while preserving data privacy and maintaining model accuracy. © 2023 IEEE.

关键词： Blockchain Differential privacy Federated learning Reputation evaluation

来源：评论

学校读者我要写书评

暂无评论

IMAN: An Adaptive Network for Robust NPC Mortality Prediction with Missing Modalities

IMAN: An Adaptive Network for Robust NPC Mortality Predictio...

引用

IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

作者： Yejing Huo Guoheng Huang Lianglun Cheng Jianbin He Xuhang Chen Xiaochen Yuan Guo Zhong Chi-Man Pun School of Computer Science and Technology Guangdong University of Technology School of Computer Science and Engineering Huizhou University Department of Computer and Information Science University of Macau Faculty of Applied Sciences Macao Polytechnic University School of Information Science and Technology Guangdong University of Foreign Studies

ISBN: (数字)9798350386226

ISBN: (纸本)9798350386233

Accurate prediction of mortality in nasopharyngeal carcinoma (NPC), a complex malignancy particularly challenging in advanced stages, is crucial for optimizing treatment strategies and improving patient outcomes. However, this predictive process is often compromised by the high-dimensional and heterogeneous nature of NPC-related data, coupled with the pervasive issue of incomplete multi-modal data, manifesting as missing radiological images or incomplete diagnostic reports. Traditional machine learning approaches suffer significant performance degradation when faced with such incomplete data, as they fail to effectively handle the high-dimensionality and intricate correlations across modalities. Even advanced multi-modal learning techniques like Transformers struggle to maintain robust performance in the presence of missing modalities, as they lack specialized mechanisms to adaptively integrate and align the diverse data types, while also capturing nuanced patterns and contextual relationships within the complex NPC data. To address these problem, we introduce IMAN: an adaptive network for robust NPC mortality prediction with missing modalities. IMAN features three integrated modules: the Dynamic Cross-Modal Calibration (DCMC) module employs adaptive, learnable parameters to scale and align medical images and field data; the Spatial-Contextual Attention Integration (SCAI) module enhances traditional Transformers by incorporating positional information within the self-attention mechanism, improving multi-modal feature integration; and the Context-Aware Feature Acquisition (CAFA) module adjusts convolution kernel positions through learnable offsets, allowing for adaptive feature capture across various scales and orientations in medical image modalities. Extensive experiments on our proprietary NPC dataset demonstrate IMAN’s robustness and high predictive accuracy, even with missing data. Compared to existing methods, IMAN consistently outperforms in scenarios with inc

关键词： Adaptive systems Accuracy Convolution Mortality Transformers Robustness Planning Medical diagnosis Kernel Medical diagnostic imaging

来源：评论

学校读者我要写书评

暂无评论

Interleaved-Modal Chain-of-Thought

arXiv

引用

arXiv 2024年

作者： Gao, Jun Li, Yongqi Cao, Ziqiang Li, Wenjie School of Computer Science and Technology Soochow University Taiwan Department of Computer Science The Hong Kong Polytechnic University Hong Kong

Chain-of-Thought (CoT) prompting elicits large language models (LLMs) to produce a series of intermediate reasoning steps before arriving at the final answer. However, when transitioning to vision-language models (VLMs), their text-only rationales struggle to express the fine-grained associations with the original image. In this paper, we propose an image-incorporated multimodal Chain-of-Thought, named Interleaved-modal Chain-of-Thought (ICoT), which generates sequential reasoning steps consisting of paired visual and textual rationales to infer the final answer. Intuitively, the novel ICoT requires VLMs to enable the generation of fine-grained interleaved-modal content, which is hard for current VLMs to fulfill. Considering that the required visual information is usually part of the input image, we propose Attention-driven Selection (ADS) to realize ICoT over existing VLMs. ADS intelligently inserts regions of the input image to generate the interleaved-modal reasoning steps with ignorable additional latency. ADS relies solely on the attention map of VLMs without the need for parameterization, and therefore it is a plug-and-play strategy that can be generalized to a spectrum of VLMs. We apply ADS to realize ICoT on two popular VLMs of different architectures. Extensive evaluations of three benchmarks have shown that ICoT prompting achieves substantial performance (up to 14%) and interpretability improvements compared to existing multimodal CoT prompting methods. © 2024, CC BY.

关键词： Visual languages

来源：评论

学校读者我要写书评

暂无评论

Investigating Object Sorting Problem for Smart Garbage Application 5

Investigating Object Sorting Problem for Smart Garbage Appli...

引用

5th International Conference on Advancements in Computational sciences, ICACS 2024

作者： Ma, Wing Shiu Yuen, Shing Hang Keung, Chi Nam Qureshi, Mehak Fatima Qureshi, Umair Mujtaba Umair, Zuneera The Hong Kong Polytechnic University School of Professional Education and Executive Development Hong Kong Mehran University of Engineering and Technology Department of Telecommunication Engineering Sindh Pakistan Chinese University of Hong Kong Kowloon Department of Computer Science & Engineering Hong Kong The Hong Kong Baptist University Department of Computer Science Hong Kong

ISBN: (纸本)9798350394764

In recent years, the rapid growth in industrialization is observed to promote the economic growth of nations. This indeed has raised the financial standards and ways of living from sustainability to a well-off living standards, but has counter effects on environmental conditions such as increasing waste generation, pollution etc, which has become a serious concern and can be alarming situation in nearby future. Thus, this research aims to create a sustainable solution to reduce the recycling material using the advance AI deep learning technology. In this research, we have designed and developed a smart recycling garbage bin that can effectively sort different types of recycling waste i.e. glass, paper, metal and plastic. To develop the smart garbage bin, the pre-trained Tensorflow model is utilized, after evaluating its performance and comparative analyses with YOLO model on the customize dataset. The final results shows that the model classified all the garbage images from the garbage used to test the smart garbage bin and achieved the high overall accuracy of 91%. © 2024 IEEE.

关键词： Recycling

来源：评论

学校读者我要写书评

暂无评论

Secrecy Analysis of Collaborative Noma Networks with Iqi

SSRN

引用

SSRN 2024年

作者： Sun, Jiangfeng Zhang, Zihan Tu, Juan Zeng, Yanyang Li, Xingwang College of Computer Science and Technology Henan Polytechnic University Jiaozuo454000 China The Wuhan Maritime Communication Research Institute Wuhan430079 China School of Physics and Electronic Information Engineering Henan Polytechnic University Jiaozuo454000 China

This paper studies the secrecy performance of collaborative non-orthogonal multiple access (NOMA) systems over k-u shadowed distribution with in-phase and quadrature-phase imbalance (IQI). The analytical expressions for outage probability (OP), intercept probability (IP), secure outage probability (SOP), and strictly positive secrecy capacity (SPSC) are derived relying on the NOMA eavesdropping model. Moreover, the Monte Carlo simulation perfectly coincides with the theoretical derivation, which confirms the accuracy of academic results. The numerical analysis indicates that IQI reduces reliability but enhances the secrecy of relaying NOMA networks. Additionally, the increase of k or the decrease of u can weaken the antijamming capability, whereas reducing m will strengthen the confidentiality. Furthermore, to realize real-time secrecy analysis, we investigate the anti-interference ability of the relaying system by the Elman neural network. Compared to other algorithms, the prediction accuracy is improved by 69.9% on average, and the runtime is decreased. © 2024, The Authors. All rights reserved.

关键词： Reliability analysis

来源：评论

学校读者我要写书评

暂无评论

GUI: A Comprehensive Dataset of Global Urban Infrastructure Based on Geospatial Visual Foundation Models 24

GUI: A Comprehensive Dataset of Global Urban Infrastructure ...

引用

32nd ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems, ACM SIGSPATIAL 2024

作者： Han, Zhenyu Zhang, Xin Xi, Yanxin Luo, Yan Xia, Tong Li, Yong Department of Electronic Engineering BNRist Tsinghua University Beijing China Shenzhen International Graduate School Tsinghua University Shenzhen China Department of Computer Science University of Helsinki Helsinki Finland Media Lab Massachusetts Institute of Technology Department of Computing The Hong Kong Polytechnic University Hong Kong Hong Kong Department of Computer Science and Technology University of Cambridge Cambridge United Kingdom

ISBN: (纸本)9798400711077

The substantial social and financial costs of infrastructure identification impede in-depth analyses of sustainable urban design, especially in developing countries. In this paper, we present a novel framework with interactive web visualization based on geospatial visual foundation models. Leveraging this framework, we examine the urban infrastructure information in 1,178 cities worldwide, covering 93, 088 km2 areas. Cross-validation reveals that the overall accuracy of identified infrastructure achieves 67.0%. It sheds light on the sustainable development of cities and exposes the stark inequity in urban infrastructure provision for vulnerable populations. The identified urban infrastructure dataset of this study are available at https://***/tsinghua-fib-lab/GUI, and the interactive web application is at https://***/yz7xbfy3. © 2024 Copyright held by the owner/author(s).

关键词： Artificial Intelligence

来源：评论

学校读者我要写书评

暂无评论

Facial expression recognition via ResNet-50

引用

International Journal of Cognitive Computing in Engineering 2021年 2卷 57-64页

作者： Li, Bin Lima, Dimas School of Computer Science and Technology Henan Polytechnic University Jiaozuo 454000 Henan China Department of Electrical Engineering Federal University of Santa Catarina Florianópolis Brazil

As one of the most important directions in the field of computer vision, facial emotion recognition plays an important role in people's daily work and life. Human emotion recognition based on facial expressions is of great significance in the application of intelligent human-computer interaction. However, in the current research on facial emotion recognition, there are some problems such as poor generalization ability of network model and low robustness of recognition system. In this content, we propose a method of feature extraction using the deep residual network ResNet-50, which combines convolutional neural network for facial emotion recognition. Through the experimental simulation of the specified data set, it can be proved that this model is superior to the current mainstream facial emotion recognition models in the performance of facial emotion detection. © 2021

关键词： Deep residual network Facial expression recognition ResNet-50

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：