检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Zhou, Yu Wu, Xingyu Wu, Jibin Feng, Liang Tan, Kay Chen Department of Computing The Hong Kong Polytechnic University Hong Kong Department of Data Science and Artificial Intelligence The Hong Kong Polytechnic University Hong Kong College of Computer Science Chongqing University Chongqing China

Model merging is a technique that combines multiple large pretrained models into a single model with enhanced performance and broader task adaptability. It has gained popularity in large pretrained model development due to its ability to bypass the need for original training data and further training processes. However, most existing model merging approaches focus solely on exploring the parameter space, merging models with identical architectures. Merging within the architecture space, despite its potential, remains in its early stages due to the vast search space and the challenges of layer compatibility. This paper marks a significant advance toward more flexible and comprehensive model merging techniques by modeling the architecture-space merging process as a reinforcement learning task. We train policy and value networks using offline sampling of weight vectors, which are then employed for the online optimization of merging strategies. Moreover, a multi-objective optimization paradigm is introduced to accommodate users’ diverse task preferences, learning the Pareto front of optimal models to offer customized merging suggestions. Experimental results across multiple tasks, including text translation, mathematical reasoning, and code generation, validate the effectiveness and superiority of the proposed framework in model merging. The code will be made publicly available after the review process. Copyright © 2024, The Authors. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Copula Guided Parallel Gibbs Sampling for Nonparametric and Coherent Topic Discovery (Extended Abstract) 39

Copula Guided Parallel Gibbs Sampling for Nonparametric and ...

引用

39th IEEE International Conference on data Engineering, ICDE 2023

作者： Lin, Lihui Rao, Yanghui Xie, Haoran Lau, Raymond Y. K. Yin, Jian Wang, Fu Lee Li, Qing Sun Yat-sen University School of Computer Science and Engineering Guangzhou China Lingnan University Department of Computing and Decision Sciences New Territories Hong Kong City University of Hong Kong Department of Information Systems Kowloon Hong Kong Sun Yat-sen University Guangdong Key Laboratory of Big Data Analysis and Processing Guangzhou China Hong Kong Metropolitan University School of Science and Technology Kowloon Hong Kong The Hong Kong Polytechnic University Department of Computing Kowloon Hong Kong

ISBN: (纸本)9798350322279

In terms of the generative process, the Gamma-Gamma-Poisson Process (G2PP) is equivalent to the nonparametric topic model of Hierarchical Dirichlet Process (HDP). Considering the high computational cost of estimating parameters in HDP, a parallel G2PP was developed to generate topics efficiently via multi-threading. Unfortunately, the above model needs to predefine the number of topics. To address this issue, we first propose a Topic Self-Adaptive Model (TSAM) for nonparametric and parallel topic discovery. In TSAM, a monitor-executor mechanism is developed to manage the global topic information using a hierarchical structure of threads. Based on the apparatus of copulas, we further extend our TSAM to TSAMcop for coherent topic modeling by exploiting a copula guided parallel Gibbs sampling algorithm. Extensive experiments validate the effectiveness of both TSAM and TSAMcop. © 2023 IEEE.

关键词： Information management

来源：评论

学校读者我要写书评

暂无评论

Emails by LLMs: A Comparison of Language in AI-Generated and Human-Written Emails 25

Emails by LLMs: A Comparison of Language in AI-Generated and...

引用

Proceedings of the 17th ACM Web science Conference 2025

作者： Weijiang Li Yinmeng Lai Sandeep Soni Koustuv Saha Siebel School of Computing and Data Science University of Illinois Urbana-Champaign Urbana IL USA Department of Quantitative Theory and Methods Emory University Atlanta GA USA

来源：评论

学校读者我要写书评

暂无评论

A Cloud computing User Experience Focused Load Balancing Method Based on Modified CMA-ES Algorithm 18th

A Cloud Computing User Experience Focused Load Balancing Met...

引用

18th International Conference on Green, Pervasive, and Cloud computing, GPC 2023

作者： Luo, Jihai Dong, Chen Chen, Zhenyi Xu, Li Chen, Tianci College of Computer and Data Science - Fuzhou University Xueyuan Road Fujian Fuzhou360108 China Fujian Key Laboratory of Network Computing and Intelligent Information Processing Fuzhou University Fuzhou China Department of Computer Science and Engineering University of South Florida TampaFL33620 United States College of Computer and Cyber Security Fujian Normal University Fuzhou350007 China

ISBN: (纸本)9789819998951

The development of the software and hardware has brought about the abundance and overflow of computing resources. Many Internet companies can lease idle computing resources based on the peak and valley cycles of usage load to provide IaaS service. After the surging development, more and more companies are gradually paying attention to the specific user experience in cloud computing. But there are few works about that aspect in load balance problem. Therefore, we consider the load balancing resource allocation problem, from the perspective of user experience, mainly based on geographical distance and regional Equilibrium, combined with user usage habits, in multiple service periods. A detailed mathematical definition has been established for this problem. This article proposes a mathematical model for the problem, along with an modified CMA-ES (Covariance Matrix Adaptation Evolution Strategy) algorithm, called WS-ESS-CMA-ES algorithm, to allocate computing resources and solve the above problem. The process of using the proposed algorithm to solve the problem of cloud computing resource allocation in real scenarios is simulated, and compared with some other algorithms. The experiment results show that our algorithm performs well. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd 2024.

关键词： Cloud computing

来源：评论

学校读者我要写书评

暂无评论

Meibomian glands segmentation in infrared images with limited annotation

引用

International Journal of Ophthalmology(English edition) 2024年第3期17卷 401-407页

作者： Jia-Wen Lin Ling-Jie Lin Feng Lu Tai-Chen Lai Jing Zou Lin-Ling Guo Zhi-Ming Lin Li Li College of Computer and Data Science Fuzhou UniversityFuzhou 350108Fujian ProvinceChina Fujian Provincial Key Laboratory of Networking Computing and Intelligent Information Processing Fuzhou UniversityFuzhou 350108Fujian ProvinceChina Shengli Clinical Medical College of Fujian Medical University Fuzhou 350001Fujian ProvinceChina Department of Ophthalmology Fujian Provincial Hospital South BranchFujian Provincial HospitalFuzhou 350001Fujian ProvinceChina

●AIM:To investigate a pioneering framework for the segmentation of meibomian glands(MGs),using limited annotations to reduce the workload on ophthalmologists and enhance the efficiency of clinical diagnosis.●METHODS:Totally 203 infrared meibomian images from 138 patients with dry eye disease,accompanied by corresponding annotations,were gathered for the study.A rectified scribble-supervised gland segmentation(RSSGS)model,incorporating temporal ensemble prediction,uncertainty estimation,and a transformation equivariance constraint,was introduced to address constraints imposed by limited supervision information inherent in scribble *** viability and efficacy of the proposed model were assessed based on accuracy,intersection over union(IoU),and dice coefficient.●RESULTS:Using manual labels as the gold standard,RSSGS demonstrated outcomes with an accuracy of 93.54%,a dice coefficient of 78.02%,and an IoU of 64.18%.Notably,these performance metrics exceed the current weakly supervised state-of-the-art methods by 0.76%,2.06%,and 2.69%,***,despite achieving a substantial 80%reduction in annotation costs,it only lags behind fully annotated methods by 0.72%,1.51%,and 2.04%.●CONCLUSION:An innovative automatic segmentation model is developed for MGs in infrared eyelid images,using scribble annotation for *** model maintains an exceptionally high level of segmentation accuracy while substantially reducing training *** holds substantial utility for calculating clinical parameters,thereby greatly enhancing the diagnostic efficiency of ophthalmologists in evaluating meibomian gland dysfunction.

关键词： infrared meibomian glands images meibomian gland dysfunction meibomian glands segmentation weak supervision scribbled annotation

来源：评论

学校读者我要写书评

暂无评论

A Survey on Usage of Multimedia databases for Emotion Elicitation: A Quantitative Report on How Content Diversity Can Improve Performance

A Survey on Usage of Multimedia Databases for Emotion Elicit...

引用

Proceedings of the International Convention MIPRO

作者： M. Horvat P. Jerčić Department of Applied Computing Faculty of Electrical Engineering and Computing University of Zagreb Zagreb Croatia Graz University of Technology Institute of Interactive Systems and Data Science Graz Austria

Affective picture databases provide a standardized set of images to elicit controlled and consistent emotional responses in research participants. They are a valuable tool for studying various emotion-related phenomena across several research domains. These domains include emotion perception, emotion regulation, and the neural basis of emotion. However, affective picture databases have diverse schemas, structures, and content, making them difficult to use. Searching and retrieving optimal pictures relevant to affective stimulation may be challenging and time-consuming. In this context, we surveyed domain experts about their practices and experiences working with affective multimedia databases such as IAPS, NAPS, OASIS, GAPED, and others. The survey identified a need for novel data observatory software. This finding motivates the authors’ intention to develop and validate such software platform that relies on AI. Such a platform would describe better, retrieve, and integrate various semi-structured affective multimedia datasets. The results prominently indicate the overwhelming dissatisfaction regarding stimuli content diversity and cultural bias, specifically regarding emotional and semantic context. The main driver of satisfaction from users of existing automated retrieval software is the quality of semantic descriptors available. This points to the direction AI should take in novel data observatory software. This survey follows up on a similar survey conducted ten years ago and explores the differences in researchers’ opinions and experiences during that time. The complete aggregated results are publicly available at https://***/mhorvat/stimdbsurvey.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Improvement of Morphological Analysis based on Three Layers of Lemmatization 24

Improvement of Morphological Analysis based on Three Layers ...

引用

10th International Conference on computing and Artificial Intelligence, ICCAI 2024

作者： Kadir, Rabiah Abdul Ting, Mary Al-Fawareh, Hejab Institute of Visual Informatics Universiti Kebangsaan Malaysia Selangor Bangi43600 Malaysia Jalan Teknologi 5 Taman Teknologi 57000 Malaysia Kuala Lumpur Wilayah Persekutuan Kuala Lumpur Malaysia Department of Computing and Data Science Xiamen University Malaysia Jalan Sunsuria Bandar Sunsuria Selangor Sepang43900 Malaysia

ISBN: (纸本)9798400717055

Currently, the world impress with the ChatGPT. ChatGPT works with more than one Natural Language Processing (NLP) techniques to analyzed and inputted data in text form. Started from analysing the lexical and the syntactic of text, it is moved to pragmatic or discourse analysis. To process unstructured data such as natural language text, the apps should start with the morphological analysis. Morphological analysis plays a critical role in NLP to study the internal structure of words that involve with syntactic and semantic analysis. Lemmatization is one of the technique in morphological analysis involves with high-level processing techniques to increase the accuracy. This paper proposes a new model that incorporates the Stanford morphology class, WordNet lexical dictionary, adaptive learning framework, and longest-match algorithm in lemmatization process. The outcome of the proposed model shows tremendous improvement with 96% accuracy compared with other existing algorithms in morphological analysis. Improvement of the morphological analysis will increase the accuracy of the semantic meaning of any natural language text processing apps like chatbot, question answering system etc. © 2024 ACM.

关键词： Lemmatization Morphology Natural Language Processing

来源：评论

学校读者我要写书评

暂无评论

Diabetic Retinopathy Imaging: Network Exudate Detection to Classify Abnormalities

Diabetic Retinopathy Imaging: Network Exudate Detection to C...

引用

Intelligent computing and Control for Engineering and Business Systems (ICCEBS)

作者： A. Chinnasamy P. Selvakumari Department of Data Science and Business Systems school of computing SRMIST kattankulathur campus Chennai Department of Computer Science and Engineering Chennai Institute of Technology Chennai Anna University

The DR becomes increasingly common, there is a need to automatically extract and classify disease severity. Due to diabetes problems, about 2% of people with this disease become completely blind, and DR complications make him 10% visually impaired if he has diabetes for 15 years. Furthermore, it plays an important role in the progression of blindness in middle-aged and older adults. If the disease is not recognized as soon as possible, patients can experience a severe stage of irreversible blindness. The shortage of ophthalmologists is a serious problem for the growing number of diabetic patients. Hard exudates must be found to screen and assist in disease monitoring and diagnosis. Therefore, Lloyd's clustering technique was used in this work.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Capturing the Complexity of Human Strategic Decision-Making with Machine Learning

arXiv

引用

arXiv 2024年

作者： Zhu, Jian-Qiao Peterson, Joshua C. Enke, Benjamin Griffiths, Thomas L. Department of Computer Science Princeton University United States Faculty of Computing and Data Science Boston University United States Department of Economics Harvard University NBER United States Department of Psychology Princeton University United States

Understanding how people behave in strategic settings–where they make decisions based on their expectations about the behavior of others–is a longstanding problem in the behavioral sciences. We conduct the largest study to date of strategic decision-making in the context of initial play in two-player matrix games, analyzing over 90,000 human decisions across more than 2,400 procedurally generated games that span a much wider space than previous datasets. We show that a deep neural network trained on these data predicts people’s choices better than leading theories of strategic behavior, indicating that there is systematic variation that is not explained by those theories. We then modify the network to produce a new, interpretable behavioral model, revealing what the original network learned about people: their ability to optimally respond and their capacity to reason about others are dependent on the complexity of individual games. This context-dependence is critical in explaining deviations from the rational Nash equilibrium, response times, and uncertainty in strategic decisions. More broadly, our results demonstrate how machine learning can be applied beyond prediction to further help generate novel explanations of complex human behavior. © 2024, CC BY.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Performance Optimization for Vehicular Cooperative Sensing: A Graph Attention Based Reinforcement Learning Approach

Performance Optimization for Vehicular Cooperative Sensing: ...

引用

2024 IEEE Global Communications Conference, GLOBECOM 2024

作者： Li, Xuefei Chen, Mingzhe Liu, Danpu Zhang, Zhilong Quek, Tony Q. S. Beijing Laboratory of Advanced Information Network Beijing University of Posts and Telecommunications Beijing100876 China University of Miami Department of Electrical and Computer Engineering Institute for Data Science and Computing Coral GablesFL33146 United States Information Systems Technology and Design Pillar Singapore University of Technology and Design 487372 Singapore

ISBN: (纸本)9798350351255

In this paper, the problem of collaborative vehicle sensing is investigated. In the considered model, a set of cooperative vehicles provide sensing information to sensing request vehicles with limited sensing and communication resources. A base station (BS) determines the subset of sensing request vehicles that each cooperative vehicle will serve and the sub-regions that each cooperative vehicle will detect. We formulate an optimization problem aiming to maximize the number of successfully detected sub-regions of sensing request vehicles while satisfying the cooperative sensing energy requirement by jointly determining the cooperative vehicle association and the sensing sub-region selection. To solve this problem, we propose a graph attention based reinforcement learning (RL) algorithm that can generate the graph information vectors based on the correlation between each cooperative vehicle and each sensing request vehicle. Using the learned graph information, the joint cooperative vehicle association and sensing sub-region selection strategy will be determined. Simulation results show that the proposed scheme can improve the number of successfully detected sub-regions of sensing request vehicles by up to 12.5% compared to the conventional RL algorithm without using graph attention networks (GANs). © 2024 IEEE.

关键词： Reinforcement 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：