检索结果-内蒙古大学图书馆

FedDAD: Federated Domain Adaptation for Object Detection

IEEE Access 2023年 11卷 51320-51330页

作者： Lu, Peggy Joy Jui, Chia-Yung Chuang, Jen-Hui National Yang Ming Chiao Tung University Department of Computer Science Hsinchu30010 Taiwan National Center for High-Performance Computing Data Science and Technology Division Taichung40763 Taiwan

Training an object detection model often requires numerous annotated images on a centralized host, which may violate user privacy and data confidentiality. Federated learning (FL) resolves this issue by allowing multiple clients, e.g., cameras, to collaboratively train a model while protecting user privacy. However, models trained with FL may fail to be generalized for new target domain due to domain shift when the data between source and target domains are statistically different. In this work, we formulate a real-world object detection problem as a source-free multi-domain adaptation problem in FL architecture. Moreover, we propose an adaptive FL algorithm, called FedDAD (Federated Domain Adaptive Detector), which aggregates models with dynamic attention targeting the unsupervised domain on server, and utilize instance-level alignment to alleviate the effects of scene variation on clients. Experimental results show that FedDAD improves the average precision (AP) by up to 10.05% and 19.15% compared to the popular FedAvg for specific object classes in the KAIST and MI3 datasets, respectively. © 2013 IEEE.

关键词： Image annotation

来源：评论

学校读者我要写书评

暂无评论

Finding NEM-U: Explaining unsupervised representation learning through neural network generated explanation masks 41

Finding NEM-U: Explaining unsupervised representation learni...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Møller, Bjørn Leth Igel, Christian Wickstrøm, Kristoffer Knutsen Sporring, Jon Jenssen, Robert Ibragimov, Bulat Department of Computer Science University of Copenhagen Copenhagen Denmark Department of Physics and Technology UiT The Arctic University of Norway Tromsø Norway Norwegian Computing Center Oslo Norway

Unsupervised representation learning has become an important ingredient of today's deep learning systems. However, only a few methods exist that explain a learned vector embedding in the sense of providing information about which parts of an input are the most important for its representation. These methods generate the explanation for a given input after the model has been evaluated and tend to produce either inaccurate explanations or are slow, which limits their practical use. To address these limitations, we introduce the Neural Explanation Masks (NEM) framework, which turns a fixed representation model into a self-explaining system by augmenting it with a masking neural network. This network provides occlusion-based explanations in parallel to computing the representations during inference. We present an instance of this framework, the NEM-U (NEM using U-net structure) architecture, which leverages similarities between segmentation and occlusion-based explanation masks. Our experiments show that NEM-U generates explanations faster and with lower complexity compared to the current state-of-the-art while maintaining high accuracy as measured by locality. Copyright 2024 by the author(s)

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Stability in Games with Uncertain Factors 16

Stability in Games with Uncertain Factors

引用

16th International Conference Management of Large-Scale System Development, MLSD 2023

作者： Gorelov, Mikhail Federal Research Center 'Computer Science and Control' The Russian Academy of Sciences Department of Information and Computing Systems Moscow Russia

The maximal guaranteed result in a hierarchical game with an undetermined factor is found in the class of strategies with feedback. The stability of the problem under consideration concerning perturbations of the payo... 详细信息

ISBN: (纸本)9798350337907

关键词： hierarchical games with uncertain factors maximal guaranteed result stability

来源：评论

学校读者我要写书评

暂无评论

Internet Representation of Research Institutions as a Result of science Reform 16

Internet Representation of Research Institutions as a Result...

引用

16th International Conference Management of Large-Scale System Development, MLSD 2023

作者： Medennikov, Viktor Federal Research Center 'Computer Science and Control' of the Russian Academy of Sciences Information and Computing Systems Department Moscow Russia

ISBN: (纸本)9798350337907

The paper considers the condition and comparison of representation on the Internet of both unaffected and reformed research institutions, in order to form a methodology for assessing the possibility of adequate and timely response to the enormous social and economic challenges occurring in the world;for this purpose, websites' content condition and dynamics are analyzed using the example of agricultural research institutions based on monitoring thereof in 2017 and 2022. It is shown that the science reform proved itself inconsistent since at first it did not respond to the demands of production during the pandemic, and now to the requests due to sanctions. The result of the science reform was a significant increase in the number of uninformative websites, which contradicts the new social order that all sectors of society, business, and management generate for digital tools for transferring scientific developments to the economy against the backdrop of severe sanctions and the pandemic. © 2023 IEEE.

关键词： Websites

来源：评论

学校读者我要写书评

暂无评论

Detecting Double Mixed Compressed Images Based on Quaternion Convolutional Neural Network

引用

Chinese Journal of Electronics 2024年第3期33卷 657-671页

作者： Hao WANG Jinwei WANG Xuelong HU Bingtao HU Qilin YIN Xiangyang LUO Bin MA Jinsheng SUN Department of Automation Nanjing University of Science and Technology Engineering Research Center of Digital Forensics Ministry of Education Department of Computer and Software Nanjing University of Information Science and Technology School of Management Nanjing University of Posts and Telecommunications Department of Computer Science and Engineering Guangdong Province Key Laboratory of Information Security Technology Sun Yat-sen University State Key Laboratory of Mathematical Engineering and Advanced Computing Qilu University of Technology

Detection of color images that have undergone double compression is a critical aspect of digital image *** the existence of various methods capable of detecting double Joint Photographic Experts Group(JPEG) compression,they are unable to address the issue of mixed double compression resulting from the use of different compression *** particular,the implementation of Joint Photographic Experts Group 2000(JPEG2000)as the secondary compression standard can result in a decline or complete loss of performance in existing *** tackle this challenge of JPEG+JPEG2000 compression,a detection method based on quaternion convolutional neural networks(QCNN) is *** QCNN processes the data as a quaternion,transforming the components of a traditional convolutional neural network(CNN) into a quaternion *** relationships between the color channels of the image are preserved,and the utilization of color information is ***,the method includes a feature conversion module that converts the extracted features into quaternion statistical features,thereby amplifying the evidence of double *** results indicate that the proposed QCNN-based method improves,on average,by 27% compared to existing methods in the detection of JPEG+JPEG2000 compression.

关键词： Color image forensics JPEG JPEG2000 Mixed double compression Quaternion convolutional neural network

来源：评论

学校读者我要写书评

暂无评论

OneBit: Towards Extremely Low-bit Large Language Models 38

OneBit: Towards Extremely Low-bit Large Language Models

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Xu, Yuzhuang Han, Xu Yang, Zonghan Wang, Shuo Zhu, Qingfu Liu, Zhiyuan Liu, Weidong Che, Wanxiang Research Center for Social Computing and Information Retrieval Harbin Institute of Technology Harbin China Department of Computer Science & Technology Tsinghua University Beijing China

Model quantification uses low bit-width values to represent the weight matrices of existing models to be quantized, which is a promising approach to reduce both storage and computational overheads of deploying highly anticipated LLMs. However, current quantization methods suffer severe performance degradation when the bit-width is extremely reduced, and thus focus on utilizing 4-bit or 8-bit values to quantize models. This paper boldly quantizes the weight matrices of LLMs to 1-bit, paving the way for the extremely low bit-width deployment of LLMs. For this target, we introduce a 1-bit model compressing framework named OneBit, including a novel 1-bit parameter representation method to better quantize LLMs as well as an effective parameter initialization method based on matrix decomposition to improve the convergence speed of the quantization framework. Sufficient experimental results indicate that OneBit achieves good performance (at least 81% of the non-quantized performance on LLaMA models) with robust training processes when only using 1-bit weight matrices. Code and checkpoints are available at https://***/xuyuzhuang11/OneBit. © 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

Automated Autism Spectral Disorder Classification Using Optimal Machine Learning Model

引用

computers, Materials & Continua 2023年第3期74卷 5251-5265页

作者： Hanan Abdullah Mengash Hamed Alqahtani Mohammed Maray Mohamed K.Nour Radwa Marzouk Mohammed Abdullah Al-Hagery Heba Mohsen Mesfer Al Duhayyim Department of Information Systems College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityP.O.Box 84428Riyadh11671Saudi Arabia Department of Information Systems College of Computer ScienceCenter of Artificial Intelligence and Unit of CybersecurityKing Khalid UniversityAbhaSaudi Arabia Department of Information Systems College of Computer ScienceKing Khalid UniversityAbhaSaudi Arabia Department of Computer Sciences College of Computing and Information SystemUmm Al-Qura UniversitySaudi Arabia Department of Computer Science College of ComputerQassim UniversitySaudi Arabia Department of Computer Science Faculty of Computers and Information TechnologyFuture University in EgyptNew Cairo11835Egypt Department of Computer Science College of Sciences and Humanities-AflajPrince Sattam bin Abdulaziz UniversitySaudi Arabia

Autism Spectrum Disorder (ASD) refers to a neuro-disorder wherean individual has long-lasting effects on communication and interaction *** information technologywhich employs artificial intelligence(AI) model has assisted in early identify ASD by using pattern *** advances of AI models assist in the automated identification andclassification of ASD, which helps to reduce the severity of the *** study introduces an automated ASD classification using owl searchalgorithm with machine learning (ASDC-OSAML) model. The proposedASDC-OSAML model majorly focuses on the identification and classificationof ASD. To attain this, the presentedASDC-OSAML model follows minmaxnormalization approach as a pre-processing stage. Next, the owl searchalgorithm (OSA)-based feature selection (OSA-FS) model is used to derivefeature subsets. Then, beetle swarm antenna search (BSAS) algorithm withIterative Dichotomiser 3 (ID3) classification method was implied for ASDdetection and classification. The design of BSAS algorithm helps to determinethe parameter values of the ID3 classifier. The performance analysis of theASDC-OSAML model is performed using benchmark dataset. An extensivecomparison study highlighted the supremacy of the ASDC-OSAML modelover recent state of art approaches.

关键词： Autism spectral disorder machine learning owl search algorithm feature selection id3 classifier

来源：评论

学校读者我要写书评

暂无评论

Exploiting a No-Regret Opponent in Repeated Zero-Sum Games

引用

Journal of Shanghai Jiaotong University (science) 2025年第2期30卷 385-398页

作者： Li, Kai Huang, Wenhan Li, Chenchen Deng, Xiaotie Department of Computer Science and Engineering Shanghai Jiao Tong University Shanghai200240 China Beijing Jingdong Century Trading Co. Ltd. Beijing100176 China Center on Frontiers of Computing Studies Peking University Beijing100871 China

In repeated zero-sum games, instead of constantly playing an equilibrium strategy of the stage game, learning to exploit the opponent given historical interactions could typically obtain a higher utility. However, when playing against a fully adaptive opponent, one would have difficulty identifying the opponent’s adaptive dynamics and further exploiting its potential weakness. In this paper, we study the problem of optimizing against the adaptive opponent who uses no-regret learning. No-regret learning is a classic and widely-used branch of adaptive learning algorithms. We propose a general framework for online modeling no-regret opponents and exploiting their weakness. With this framework, one could approximate the opponent’s no-regret learning dynamics and then develop a response plan to obtain a significant profit based on the inferences of the opponent’s strategies. We employ two system identification architectures, including the recurrent neural network (RNN) and the nonlinear autoregressive exogenous model, and adopt an efficient greedy response plan within the framework. Theoretically, we prove the approximation capability of our RNN architecture at approximating specific no-regret dynamics. Empirically, we demonstrate that during interactions at a low level of non-stationarity, our architectures could approximate the dynamics with a low error, and the derived policies could exploit the no-regret opponent to obtain a decent utility. © Shanghai Jiao Tong University 2023.

关键词： Dynamical systems

来源：评论

学校读者我要写书评

暂无评论

RAUIE: A Relation-Augmented Document-Level Event Extraction Model Based on UIE 5

RAUIE: A Relation-Augmented Document-Level Event Extraction ...

引用

5th International Seminar on Artificial Intelligence, Networking and Information Technology, AINIT 2024

作者： Liang, Yiyuan Jia, Hong Wang, Licai Huang, Yangchen Yang, Zhe Big Data Research and Development Center North China Institute of Computing Technology Beijing China Tsinghua University Department of Computer Science and Technology Beijing China The Chinese University of Hong Kong Department of Computer Science and Engineering Hong Kong

ISBN: (纸本)9798350385557

In document-level event extraction tasks, multiple event types often coexist within a document (the multi-event problem), and arguments may play different roles across different events (the argument intertwining problem). Existing methods often suffer from significantly reduced event extraction accuracy, mainly when focusing on scenarios with scarce annotated data and imbalanced event-type distributions. This paper emphasizes enhancing the correlation between event triggers and arguments to address these challenges. Therefore, we propose the Relation-Augmented UIE Model (RADIE), which introduces entity relationships into prompts using prompt-based learning and designs a relation-augmentation algorithm to improve the accuracy and completeness of event extraction. Experimental results demonstrate that this approach significantly outperforms baseline methods, mainly showcasing excellent performance on datasets with long-tail distributions. © 2024 IEEE.

关键词： document-level event extraction relation augment Universal Information Extraction

来源：评论

学校读者我要写书评

暂无评论

Medical image registration and its application in retinal images:a review

引用

visual computing for Industry,Biomedicine,and Art 2024年第1期7卷 142-164页

作者： Qiushi Nie Xiaoqing Zhang Yan Hu Mingdao Gong Jiang Liu Research Institute of Trustworthy Autonomous Systems and Department of Computer Science and Engineering Southern University of Science and TechnologyShenzhen 518055China Center for High Performance Computing and Shenzhen Key Laboratory of Intelligent Bioinformatics Shenzhen Institute of Advanced TechnologyChinese Academy of SciencesShenzhen 518055China Singapore Eye Research Institute Singapore 169856Singapore State Key Laboratory of Ophthalmology Optometry and Visual ScienceEye HospitalWenzhou Medical UniversityWenzhou 325027China

Medical image registration is vital for disease diagnosis and treatment with its ability to merge diverse informa-tion of images,which may be captured under different times,angles,or *** several surveys have reviewed the development of medical image registration,they have not systematically summarized the existing med-ical image registration *** this end,a comprehensive review of these methods is provided from traditional and deep-learning-based perspectives,aiming to help audiences quickly understand the development of medical image *** particular,we review recent advances in retinal image registration,which has not attracted much *** addition,current challenges in retinal image registration are discussed and insights and prospects for future research provided.

关键词： computer-aided diagnosis Medical image registration Deep learning Generative model Transformer Retina

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：