检索结果-内蒙古大学图书馆

Beyond pixels:Learning from multimodal hyperspectral superpixels for land cover classification

science China(Technological sciences) 2022年第4期65卷 802-808页

作者： HONG DanFeng WU Xin YAO Jing ZHU XiaoXiang Key Laboratory of Computational Optical Imaging Technology Aerospace Information Research InstituteChinese Academy of SciencesBeijing 100094China School of Computer Science(National Pilot Software Engineering School) Beijing University of Posts and TelecommunicationsBeijing 100876China Remote Sensing Technology Institute German Aerospace Center(DLR)Wessling 82234Germany Data Science in Earth Observation Technical University of MunichMunich 80333Germany

Despite tons of advanced classification models that have recently been developed for the land cover mapping task,the monotonicity of a single remote sensing data source,such as only using hyperspectral data or multispectral data,hinders the classification accuracy from being further improved and tends to meet the performance *** this reason,we develop a novel superpixel-based subspace learning model,called Supace,by jointly learning multimodal feature representations from HS and MS superpixels for more accurate LCC *** can learn a common subspace across multimodal RS data,where the diverse and complementary information from different modalities can be better combined,being capable of enhancing the discriminative ability of to-be-learned features in a more effective *** better capture semantic information of objects in the feature learning process,superpixels that beyond pixels are regarded as the study object in our Supace for *** experiments have been conducted on two popular hyperspectral and multispectral datasets,demonstrating the superiority of the proposed Supace in the land cover classification task compared with several well-known baselines related to multimodal remote sensing image feature learning.

关键词： classification hyperspectral image land cover multimodal multispectral image remote sensing subspace learning superpixels

来源：评论

学校读者我要写书评

暂无评论

Flatness-aware Sequential Learning Generates Resilient Backdoors

arXiv

引用

arXiv 2024年

作者： Pham, Hoang Ta, The-Anh Tran, Anh Doan, Khoa D. College of Engineering and Computer Science VinUniversity Viet Nam CSIRO’s Data61 Australia VinAI Research Viet Nam

Recently, backdoor attacks have become an emerging threat to the security of machine learning models. From the adversary’s perspective, the implanted backdoors should be resistant to defensive algorithms, but some recently proposed fine-tuning defenses can remove these backdoors with notable efficacy. This is mainly due to the catastrophic forgetting (CF) property of deep neural networks. This paper counters CF of backdoors by leveraging continual learning (CL) techniques. We begin by investigating the connectivity between a backdoored and fine-tuned model in the loss landscape. Our analysis confirms that fine-tuning defenses, especially the more advanced ones, can easily push a poisoned model out of the backdoor regions, making it forget all about the backdoors. Based on this finding, we re-formulate backdoor training through the lens of CL and propose a novel framework, named Sequential Backdoor Learning (SBL), that can generate resilient backdoors. This framework separates the backdoor poisoning process into two tasks: the first task learns a backdoored model, while the second task, based on the CL principles, moves it to a backdoored region resistant to fine-tuning. We additionally propose to seek flatter backdoor regions via a sharpness-aware minimizer in the framework, further strengthening the durability of the implanted backdoor. Finally, we demonstrate the effectiveness of our method through extensive empirical experiments on several benchmark datasets in the backdoor domain. The source code is available at https://***/mail-research/SBL-resilient-backdoors © 2024, CC0.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

A Glimpse to Scalable LLM Architectures: Building Real-Time Sentiment Analyser with Kafka and FastAPI

A Glimpse to Scalable LLM Architectures: Building Real-Time ...

引用

International Conference on computer science and engineering (UBMK)

作者： Uzay Çetin Yunus Emre Gündoğmuş Department Summarify Computer Engineering Galatasaray University Istanbul Türkiye Growth Data Science Summarify Hepsiburada Istanbul Türkiye

ISBN: (数字)9798350365887

ISBN: (纸本)9798350365894

In today's digital world, the real-time generation and analysis of data, particularly on social media platforms, are becoming increasingly important. This study aims to design and evaluate a scalable architecture for a scenario where multiple users simultaneously perform sentiment analysis. Two different architectures were examined in the study: the Basic Architecture and the Proposed Architecture. The Basic Architecture processes and responds to user requests directly using FastAPI and Gunicorn. In contrast, the Proposed Architecture employs FastAPI, Kafka, and Redis to create a more complex and distributed structure. Test results demonstrate that the Proposed Architecture offers higher throughput and lower response times under high demand and user loads. This study presents significant findings on how to establish a scalable and high-performance infrastructure for large language models.

关键词： computer science Sentiment analysis Social networking (online) Large language models Computational modeling computer architecture Throughput Real-time systems Time factors Load modeling

来源：评论

学校读者我要写书评

暂无评论

基于数据挖掘技术的税务风险检测方法评述

引用

engineering 2024年第3期34卷 43-59页

作者： Qinghua Zheng Yiming Xu Huixiang Liu Bin Shi Jiaxiang Wang Bo Dong School of Computer Science and Technology Xi’an Jiaotong UniversityXi’an 710049China Shaanxi Provincial Key Laboratory of Big Data Knowledge Engineering Xi’an Jiaotong UniversityXi’an 710049China School of Distance Education Xi’an Jiaotong UniversityXi’an 710049China

Tax risk behavior causes serious loss of fiscal revenue,damages the country’s public infrastructure,and disturbs the market economic order of fair *** recent years,tax risk detection,driven by information technology such as data mining and artificial intelligence,has received extensive *** promote the high-quality development of tax risk detection methods,this paper provides the first comprehensive overview and summary of existing tax risk detection methods *** specifi-cally,it first discusses the causes and negative impacts of tax risk behaviors,along with the development of tax risk *** then focuses on data-mining-based tax risk detection methods utilized around the *** on the different principles employed by the algorithms,existing risk detection methods can be divided into two categories:relationship-based and non-relationship-based.A total of 14 risk detection methods are identified,and each method is thoroughly explored and ***,four major technical bottlenecks of current data-driven tax risk detection methods are analyzed and discussed,including the difficulty of integrating and using fiscal and tax fragmented knowledge,unexplainable risk detection results,the high cost of risk detection algorithms,and the reliance of existing algorithms on labeled *** investigating these issues,it is concluded that knowledge-guided and datadriven big data knowledge engineering will be the development trend in the field of tax risk in the future;that is,the gradual transition of tax risk detection from informatization to intelligence is the future development direction.

关键词： Tax risk detection data mining Knowledge guide Informatization Intellectualization

来源：评论

学校读者我要写书评

暂无评论

Transformer-Based Federated Learning For Multi-Label Remote Sensing Image Classification

Transformer-Based Federated Learning For Multi-Label Remote ...

引用

IEEE International Symposium on Geoscience and Remote Sensing (IGARSS)

作者： Barış Büyüktaş Kenneth Weitzel Sebastian Völkers Felix Zailskas Begüm Demir BIFOLD - Berlin Institute for the Foundations of Learning and Data Germany Faculty of Electrical Engineering and Computer Science Technische Universität Berlin Germany

ISBN: (数字)9798350360325

ISBN: (纸本)9798350360332

Federated learning (FL) aims to collaboratively learn deep learning model parameters from decentralized data archives (i.e., clients) without accessing training data on clients. However, the training data across clients might be not independent and identically distributed (non-IID), which may result in difficulty in achieving optimal model convergence. In this work, we investigate the capability of state-of-the-art transformer architectures (which are MLP-Mixer, ConvMixer, PoolFormer) to address the challenges related to non-IID training data across various clients in the context of FL for multi-label classification (MLC) problems in remote sensing (RS). The considered transformer architectures are compared among themselves and with the ResNet-50 architecture in terms of their: 1) robustness to training data heterogeneity; 2) local training complexity; and 3) aggregation complexity under different non-IID levels. The experimental results obtained on the BigEarthNet-S2 benchmark archive demonstrate that the considered architectures increase the generalization ability with the cost of higher local training and aggregation complexities. On the basis of our analysis, some guidelines are derived for a proper selection of transformer architecture in the context of FL for RS MLC. The code of this work is publicly available at https://***/rsim/FL-Transformer.

关键词： Training Accuracy Federated learning Training data Transformers Robustness Complexity theory

来源：评论

学校读者我要写书评

暂无评论

Towards Portable Identities in the Matrix Protocol 42

Towards Portable Identities in the Matrix Protocol

引用

42nd IEEE International Conference on Distributed Computing Systems Workshops, ICDCSW 2022

作者： Ihle, Cornelius Deifub, Fabian Schubotz, Moritz Gipp, Bela University of Göttingen Department of Computer Science Göttingen Germany Data & Knowledge Engineering University of Wuppertal Wuppertal Germany Mathematics Fiz Karlsruhe Berlin Germany

ISBN: (数字)9781665488792

ISBN: (纸本)9781665488792

In this paper we present a cryptographic challenge-response authentication mechanism to enable portable identities within the Matrix protocol. Online instant communication systems like Matrix connect people in a convenient and cost-effective manner. However, most of today's communication infrastructure relies on inherently centralized infrastructure. Matrix currently, takes a federated approach instead. However, to fully decentralize the Matrix network, further actions must take place. Among others, a user's identity must be disconnected from the server they used to sign up and happen to communicate on. As part of this effort, a challenge-response authentication mechanism has been implemented that allows user IDs to be usable on any server, even if the a user's homeserver becomes unreachable. © 2022 IEEE.

关键词： Authentication

来源：评论

学校读者我要写书评

暂无评论

Challenges in Fuzzy Decision Making for Future Research 21

Challenges in Fuzzy Decision Making for Future Research

引用

21st International Conference on New Trends in Intelligent Software Methodologies, Tools and Techniques, SoMeT 2022

作者： Trillo, José Ramón Herrera-Viedma, Enrique Higueras-Ruiz, María José Alonso, Sergio Morente-Molinera, Juan Antonio Cabrerizo, Francisco Javier Department of Computer Science and Artificial Intelligence Andalusian Research Institute in Data Science and Computational Intelligence DaSCI University of Granada Granada18071 Spain Department of Information and Communication University of Granada Granada18071 Spain Department of Software Engineering Andalusian Research Institute in Data Science and Computational Intelligence DaSCI University of Granada Granada18071 Spain

ISBN: (纸本)9781643683164

Fuzzy Decision Making is a line of research that has had an important presence in the literature since 1965 until nowadays. From that year to the present, the line of research has been consolidated. The research field goals have evolved along with all the research that makes use of these systems and the new technologies requirements. Therefore, new challenges, different from the ones that were faced at the beginning, need to be solved. The aim of this paper is to describe the recent challenges faced by those working in this line of research. This paper will help researchers to know how their work should be oriented in the future. © 2022 The authors and IOS Press. All rights reserved.

关键词： Decision making

来源：评论

学校读者我要写书评

暂无评论

Contract-Inspired Contest Theory for Controllable Image Generation in Mobile Edge Metaverse

引用

IEEE Transactions on Mobile Computing 2025年

作者： Liu, Guangyuan Du, Hongyang Wang, Jiacheng Niyato, Dusit Kim, Dong In Nanyang Technological University College of Computing and Data Science Energy Research Institute @ NTU Interdisciplinary Graduate Program Singapore University of Hong Kong Department of Electrical and Electronic Engineering Hong Kong Hong Kong Nanyang Technological University College of Computing and Data Science Singapore Sungkyunkwan University Department of Electrical and Computer Engineering Korea Republic of

The rapid advancement of immersive technologies has propelled the development of the Metaverse, where the convergence of virtual and physical realities necessitates the generation of high-quality, photorealistic images to enhance user experience. However, generating these images, especially through Generative Diffusion Models (GDMs), in mobile edge computing environments presents significant challenges due to the limited computing resources of edge devices and the dynamic nature of wireless networks. This paper proposes a novel framework that integrates contract-inspired contest theory, Deep Reinforcement Learning (DRL), and GDMs to optimize image generation in these resource-constrained environments. The framework addresses the critical challenges of resource allocation and semantic data transmission quality by incentivizing edge devices to efficiently transmit high-quality semantic data, which is essential for creating realistic and immersive images. The use of contest and contract theory ensures that edge devices are motivated to allocate resources effectively, while DRL dynamically adjusts to network conditions, optimizing the overall image generation process. Experimental results demonstrate that the proposed approach not only improves the quality of generated images but also achieves superior convergence speed and stability compared to traditional methods. This makes the framework particularly effective for optimizing complex resource allocation tasks in mobile edge Metaverse applications, offering enhanced performance and efficiency in creating immersive virtual environments. © 2002-2012 IEEE.

关键词： Resource allocation

来源：评论

学校读者我要写书评

暂无评论

HyCoT: A Transformer-Based Autoencoder for Hyperspectral Image Compression

HyCoT: A Transformer-Based Autoencoder for Hyperspectral Ima...

引用

Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing, WHISPERS

作者： Martin Hermann Paul Fuchs Behnood Rasti Begüm Demir Faculty of Electrical Engineering and Computer Science Technische Universität Berlin Germany BIFOLD - Berlin Institute for the Foundations of Learning and Data Germany

ISBN: (数字)9798331513139

ISBN: (纸本)9798331513146

The development of learning-based hyperspectral image (HSI) compression models has recently attracted significant interest. Existing models predominantly utilize convolutional filters, which capture only local dependencies. Furthermore, they often incur high training costs and exhibit substantial computational complexity. To address these limitations, in this paper we propose Hyperspectral Compression Trans-former (HyCoT) that is a transformer-based autoencoder for pixelwise HSI compression. Additionally, we apply a simple yet effective training set reduction approach to accelerate the training process. Experimental results on the HySpecNet-11k dataset demonstrate that HyCoT surpasses the state of the art across various compression ratios by over 1 dB of PSNR with significantly reduced computational require-ments. Our code and pre-trained weights are publicly avail-able at https://***/rsim/hycot.

关键词： Training Convolutional codes Image coding Convolution Computational modeling Autoencoders Transformers Sensors Computational complexity Hyperspectral imaging

来源：评论

学校读者我要写书评

暂无评论

Enhancing Convergence Efficiency in Federated Learning Models 25

Enhancing Convergence Efficiency in Federated Learning Model...

引用

25th International Arab Conference on Information Technology, ACIT 2024

作者： Samara, Ghassan Elhilo, Ala'Mazen Aljaidi, Mohammad Alamleh, Amneh Al-Mousa, Mohammad Rasmi Batyha, Radwan Alazaidah, Raed Bani-Bakr, Alaa Elrashidi, Ali Zarqa University Faculty of It Department of Computer Science Zarqa Jordan Zarqa University Faculty of Information Technology Zarqa Jordan Zarqa University Faculty of Information Technology Department of Data Science and Ai Zarqa Jordan Zarqa University Department of Cybersecurity Zarqa Jordan Applied Science Private University Faculty of Information Technology Amman Jordan Zarqa University Faculty of It Department of Cyber Security Zarqa Jordan University of Business and Technology Department of Electrical Engineering Jeddah21432 Saudi Arabia

ISBN: (纸本)9798331540012

This study evaluates federated learning models' performance compared to centralized models using the MNIST and CIFAR-10 datasets. Four models were tested: two Federated Learning models and two centralized models. The evaluation focused on accuracy, training rounds, and CO2 emissions, to optimize convergence and reducing energy consumption while maintaining strong model performance. The results showed that Federated learning models outperformed centralized models, with the Federated Learning MNIST model achieving 98.79% accuracy compared to 87.93% for the centralized model. Also, the accuracy of the Federated Learning CIFAR-10 model achieved 91.20% accuracy, exceeding the centralized model's 90.14%. The most accuracy improvements happened in the early training rounds. An adjusted hill-climbing technique was used to the centralized models, decreasing the number of iterations and CO2 emissions without significantly reducing accuracy. © 2024 IEEE.

关键词： Adversarial machine 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：