检索结果-内蒙古大学图书馆

Multi-view prototype-based disambiguation for partial label learning

PATTERN RECOGNITION 2023年第1期141卷

作者： Sun, Shiding Yu, Xiaotong Tian, Yingjie Univ Chinese Acad Sci Sch Math Sci Beijing 100049 Peoples R China Chinese Acad Sci Res Ctr Fictitious Econ & Data Sci Beijing 100190 Peoples R China Chinese Acad Sci Key Lab Big Data Min & Knowledge Management Beijing 100190 Peoples R China Univ Chinese Acad Sci Sch Econ & Management Beijing 100190 Peoples R China UCAS MOE Social Sci Lab Digital Econ Forecasts & Policy Beijing 100190 Peoples R China

In this work, we study the multi-view partial label learning (MVPLL) problem, where each instance is depicted by different view features and associated with a set of candidate labels, among which a true label exists but is inaccessible in the training phase. Most existing PLL methods only consider single view case, which learn view classifier independently and neglect the view correlations, thus can not be applied to solve MVPLL problem. Due to the non-deep framework, traditional MVPLL approach is weak in the representation ability, so its performance is still to be improved. To solve the MVPLL problem, a deep multi-view prototype-based disambiguation approach is proposed in this paper. Specifically, we innovatively employ the deep neural network for multi-view ambiguously-labeled image classification to enhance the representation ability, which makes use of the information fusion between multiple views. To improve the discriminative ability, we propose multi-view prototype-based label disambiguation algorithm. On theoretical aspect, an estimation error bound for view-risk estimator is established, which is shown to be larger than that for fuse-risk estimator. Experiments demonstrate the superiorities of our proposed method in terms of the prediction accuracy. & COPY;2023 Elsevier Ltd. All rights reserved.

关键词： Multi-view learning Partial label learning Weakly supervised learning

来源：评论

学校读者我要写书评

暂无评论

Exploiting sample correlation for crowd counting with multi-expert network 18

Exploiting sample correlation for crowd counting with multi-...

引用

18th IEEE/CVF International Conference on Computer Vision (ICCV)

作者： Liu, Xinyan Li, Guorong Han, Zhenjun Zhang, Weigang Yang, Yifan Huang, Qingming Sebe, Nicu Univ Chinese Acad Sci Beijing Peoples R China UCAS Key Lab Big Data Min & Knowledge Management Beijing Peoples R China Harbin Inst Technol Weihai Peoples R China Chinese Acad Sci Key Lab Intelligent Informat Proc ICT Beijing Peoples R China Univ Trento Trento Italy

ISBN: (纸本)9781665428125

Crowd counting is a difficult task because of the diversity of scenes. Most of the existing crowd counting methods adopt complex structures with massive backbones to enhance the generalization ability. Unfortunately, the performance of existing methods on large-scale data sets is not satisfactory. In order to handle various scenarios with less complex network, we explored how to efficiently use the multi-expert model for crowd counting tasks. We mainly focus on how to train more efficient expert networks and how to choose the most suitable expert. Specifically, we propose a task-driven similarity metric based on sample's mutual enhancement, referred as co-fine-tune similarity, which can find a more efficient subset of data for training the expert network. Similar samples are considered as a cluster which is used to obtain parameters of an expert. Besides, to make better use of the proposed method, we design a simple network called FPN with Deconvolution Counting Network, which is a more suitable base model for the multi-expert counting network. Experimental results show that multiple experts FDC (MFDC) achieves the best performance on four public data sets, including the large scale NWPU-Crowd data set. Furthermore, the MFDC trained on an extensive dense crowd data set can generalize well on the other data sets without extra training or fine-tuning.(1)

关键词： Complex networks

来源：评论

学校读者我要写书评

暂无评论

A local search algorithm with movement gap and adaptive configuration checking for the maximum weighted s-plex problem

引用

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE 2024年 133卷

作者： Li, Ruizhi Xu, Ruiqi Wang, Fangzhou Hu, Shuli Wang, Yiyuan Yin, Minghao Li, Hui Jilin Univ Finance & Econ Sch Management Sci & Informat Engn Changchun 130117 Peoples R China Northeast Normal Univ Sch Informat Sci & Technol Changchun 130117 Peoples R China Business Big Data Res Ctr Jilin Prov Changchun 130117 Peoples R China Jilin Univ Key Lab Symbol Computat & Knowledge Engn Minist Educ Changchun 130012 Peoples R China Northeast Normal Univ Key Lab Appl Stat MOE Changchun 130117 Peoples R China

The maximum weighted s-plex problem (MWSP) is an important generalization of the maximum s-plex problem and has a wide range of applications in many fields such as social network analysis. In this paper, we propose an efficient local search algorithm based on movement gap and adaptive configuration checking. Firstly, we use flip to construct the initial solution and select more suitable vertices to jump out of the local optimum. Secondly, we propose a novel variant of configuration checking with stronger automatic adjustment capability, called adaptive configuration checking (ACC), which can effectively avoid the cycling problem in local search. Besides, we use three operators Add, Swap, and Drop to improve the solution. Our algorithm determines the operator by the increment and judges which vertices are more likely to appear in the solution based on the movement gap (MG) of each vertex. In the perturbation phase, we use the Jump operator to escape the local optimum and adopt the best from multiple selections based on the dynamic parameter (DPBMS) heuristic to improve the search efficiency. By comparing the current state-of-the-art algorithms, we demonstrate that our method performs well. In addition, we validate the effectiveness of the ACC and MG strategies throughout the algorithm and the role of the DPBMS heuristic in the perturbation procedure.

关键词： Combinatorial optimization Maximum weighted s-plex problem Local search Adaptive configuration checking Movement gap

来源：评论

学校读者我要写书评

暂无评论

Analysis of slump and surge phenomenon in Chinese stock market based on sequence alignment method

引用

SOFT COMPUTING 2020年第23期24卷 18185-18202页

作者： Long, Wen Song, Linqiu Tian, Yingjie Yang, Wenning Chinese Acad Sci Res Ctr Fictitious Econ & Data Sci Beijing 100190 Peoples R China Univ Chinese Acad Sci Sch Econ & Management Beijing 100190 Peoples R China Chinese Acad Sci Key Lab Big Data Min & Knowledge Management Beijing 100190 Peoples R China

The method of sequence alignment in bioinformatics is introduced into financial time series analysis, which can capture the large-scale features of the variables, suppress the noise and reveal the implicit mode of the system from different angles without strict assumptions. Based on existing methods of sequence alignment, we propose a new scoring matrix construction method for financial sequence alignment, a purpose-oriented matrix. This method can be used to extract the feature segments of the sequence. An empirical study of the stocks of Shanghai Stock Exchange and Shenzhen Stock Exchange is implemented with this new method. The results reflect the relationship between the two stock markets under the conditions of boom or bust and confirm the feasibility and effectiveness of introducing the method into the financial analysis.

关键词： Sequence alignment Stock market Financial time series Correlation Scoring matrix

来源：评论

学校读者我要写书评

暂无评论

A Study on The Multi-Attribute Decision Theory and Methods

引用

Procedia Computer Science 2022年 214卷 544-551页

作者： Jinyuan Feng Yunxi Yan Manling Huang Yang Du Zhichen Lu Biao Li School of Business Administration Faculty of Business Administration Southwestern University of Finance and Economics Chengdu 611130 China College of Dance Minzu University of China Beijing China the School of Economics and Management University of Chinese Academy of Sciences Beijing 101408 China the Research Center on Fictitious Economy and Data Science and the Key Laboratory of Big Data Mining and Knowledge Management Chinese Academy of Sciences Beijing 100190 China

Multi-attribute decision-making problem (MADM) is a kind of multi-criteria decision-making (MCDM), which is often used to solve decision-making problems where the decision variables are discrete and the number of decision-making schemes is limited. This paper reviews the basic definition of multi-attribute decision making (MADM) through the performance evaluation of science and technology funds in a city, and then deduces the TODIM decision making method and TOPSIS decision making method in detail. Finally, the clue case is decided by combining these two decision making methods. This paper specifically shows the usage scenarios and limitations of the TODIM decision method and the TOPSIS decision method through the clue case, analyzes the advantages and disadvantages of the two methods through the comparison of the two methods, and puts forward suggestions for the clue case based on the results of the two schemes.

关键词： Multi-attribute decision-making multi-criteria decision-making TODIM TOPSIS

来源：评论

学校读者我要写书评

暂无评论

Learning Unified Embeddings for Recommendation via Meta-path Semantics 21

Learning Unified Embeddings for Recommendation via Meta-path...

引用

29th ACM International Conference on Multimedia, MM 2021

作者： Hao, Qianxiu Xu, Qianqian Yang, Zhiyong Huang, Qingming Key Laboratory of Intelligent Information Processing Institute of Computing Technology CAS Beijing China School of Computer Science and Technology University of Chinese Academy of Sciences Beijing China Key Laboratory of Big Data Mining and Knowledge Management Chinese Academy of Sciences Beijing China Artificial Intelligence Research Center Peng Cheng Laboratory Shenzhen China

ISBN: (纸本)9781450386517

Heterogeneous information networks (HINs) have become a popular tool to capture complicated user-item relationships in recommendation problems in recent years. As a typical instantiation of HINs, meta-path is introduced in search of higher-level representations of user-item interactions. Though remarkable success has been achieved along this direction, existing meta-path-based recommendation methods face at least one of the following issues: 1) existing methods merely adopt simple meta-path fusion rules, which might be insufficient to exclude inconsistent information of different meta-paths that may hurt model performance;2) the representative power is limited by shallow/stage-wise formulations. To solve these issues, we propose an end-to-end and unified embedding-based recommendation framework with graph-based learning. To address 1), we propose a flexible fusion module to integrate meta-path-based similarities into relative similarities between users and items. To address 2), we take advantage of the powerful representative ability of deep neural networks to learn more complicated and flexible latent embeddings. Finally, empirical studies on real-world datasets demonstrate the effectiveness of our proposed method. © 2021 ACM.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

ReconBoost: Boosting Can Achieve Modality Reconcilement

arXiv

引用

arXiv 2024年

作者： Hua, Cong Xu, Qianqian Bao, Shilong Yang, Zhiyong Huang, Qingming Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing China School of Computer Science and Technology University of Chinese Academy of Sciences Beijing China Institute of Information Engineering Chinese Academy of Sciences Beijing China School of Cyber Security University of Chinese Academy of Sciences Beijing China Key Laboratory of Big Data Mining and Knowledge Management Chinese Academy of Sciences Beijing China

This paper explores a novel multi-modal alternating learning paradigm pursuing a reconciliation between the exploitation of uni-modal features and the exploration of cross-modal interactions. This is motivated by the fact that current paradigms of multi-modal learning tend to explore multi-modal features simultaneously. The resulting gradient prohibits further exploitation of the features in the weak modality, leading to modality competition, where the dominant modality overpowers the learning process. To address this issue, we study the modality-alternating learning paradigm to achieve reconcilement. Specifically, we propose a new method called ReconBoost to update a fixed modality each time. Herein, the learning objective is dynamically adjusted with a reconcilement regularization against competition with the historical models. By choosing a KL-based reconcilement, we show that the proposed method resembles Friedman's Gradient-Boosting (GB) algorithm, where the updated learner can correct errors made by others and help enhance the overall performance. The major difference with the classic GB is that we only preserve the newest model for each modality to avoid overfitting caused by ensembling strong learners. Furthermore, we propose a memory consolidation scheme and a global rectification scheme to make this strategy more effective. Experiments over six multi-modal benchmarks speak to the efficacy of the method. We release the code at https://***/huacong/ReconBoost. Copyright © 2024, The Authors. All rights reserved.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Investigating Laws of Intelligence Based on AI IQ Research

引用

Annals of data Science 2020年第3期7卷 399-416页

作者： Liu, Feng Shi, Yong Research Center on Fictitious Economy and Data Science The Chinese Academy of Sciences Beijing100190 China The Key Laboratory of Big Data Mining and Knowledge Management Chinese Academy of Sciences Beijing100190 China The College of Information Science and Technology University of Nebraska at Omaha OmahaNE68118 United States

Based on the authors’ previous work on the research on the evaluation of artificial intelligence (AI) intelligence quotient and the Standard Intelligent Model, this paper proposes three laws of intelligence for interpreting the concepts of intelligence, wisdom, consciousness, life and non-life. The first law is called "M Law of Intelligence" where any Agent can be regarded as a system that has the abilities to input, output, storage (master) and creative (innovate) knowledge. The second law is called "Ω Law of Intelligence" in which any Agent is affected by new "forces" FΩ and Fα derived in this paper, which evolve to Ωpoint whose knowledge processing ability is infinite, or converge to αpoint whose knowledge processing ability is zero. The third law is called "Α Law of Intelligence", that is, when an Agent changes surrounding αpoint, the entire Universe relative to the Agent will also change between infinity and zero. The Three Laws need to be validated by a biochemical experiment method, an AI system intelligence evaluation experiment method or the computer program simulation experiment method. © 2020, Springer-Verlag GmbH Germany, part of Springer Nature.

关键词： Software engineering

来源：评论

学校读者我要写书评

暂无评论

knowledge spillover driven by institutions: evidence from the big science project in China

引用

JOURNAL OF knowledge management 2021年第1期25卷 48-84页

作者： Zhang, Lingling Gao, Chang Nakamori, Yoshiteru Univ Chinese Acad Sci Sch Econ & Management Beijing Peoples R China Chinese Acad Sci Key Lab Big Data Min & Knowledge Management Beijing Peoples R China Univ Chinese Acad Sci Sino Danish Coll Beijing Peoples R China Japan Adv Inst Sci & Technol Nomi Japan

Purpose This study aims to explore the knowledge spillover mechanism in big science projects (BSP) from an institutional perspective by elaborating on the dynamic relationship between institutional dualism and legitimacy. Design/methodology/approach The study conducts an exploratory research and adopts the grounded theory methodology in the context of BSP. data draw mainly upon nine semi-structured interviews. Findings The knowledge spillovers in BSP are driven by institutions, which work through mechanisms of legitimacy perception. Formal and informal institutions influence organizational and individual behavior through legitimacy pressure and support. Formal institutions impose legitimacy pressure on organizations and individuals, forcing them to cooperate closely to solve problems;informal institutions enable them to adopt innovative strategies and positive attitudes through legitimacy support;all these promote knowledge spillovers in research and development (R&D) activities, engineering practice and regional interaction. knowledge spillovers enable stakeholders to realize their R&D advancement, manufacturing promotion and management sophistication. Further, regional knowledge diffusion and culture transmission promote regional innovation and social capital accumulation. Research limitations/implications The study develops a theoretical model that shows how knowledge spillover mechanisms happen in BSP from an institutional perspective (the trigger, the channels/process and the impacts). More specifically, this explanation is provided by explaining how formal and informal institutions influence organizational and individual behavior through legitimacy perceptions. Practical implications First, policymakers should recognize and value the guiding, supporting and coordinating role of formal institutions and enrich capital forms to release the legitimacy pressure of stakeholders. Second, management of BSP needs to be capable of coordinating stakeholder relationships

关键词： Legitimacy knowledge spillover Institutions Grounded theory big science project University-industry-government (UIG)

来源：评论

学校读者我要写书评

暂无评论

An intelligent recommendation method for coal mine accident case via ontology knowledge service

ENGINEERING REPORTS

引用

ENGINEERING REPORTS 2024年第1期6卷

作者： Zhao, Zhigang Luo, Xiong Zhang, Xinju Chen, Maojian Univ Sci & Technol Beijing Sch Comp & Commun Engn Beijing 100083 Peoples R China Univ Sci & Technol Beijing Shunde Innovat Sch Foshan Peoples R China Beijing Key Lab Knowledge Engn Mat Sci Beijing Peoples R China Minist Emergency Management Peoples Republ China Big Data Ctr Beijing Peoples R China

Coal mine accidents, for example, water leakage and collapse, can damage the circuit system, and they in turn can affect the stable operation of the coal mine. Therefore, it is necessary to identify the causes of coal mine accidents and reduce the number of accidents in coal mines. Government and enterprises have already accumulated numerous coal mine accident cases. To summarize the characteristics and rules of accidents, survey the deep cause of the accident, avoid the recurrence of similar accidents, and early pre-control of risk source, there is a need to analyze the correlation in the accident characteristics. Then, with the help of ontology knowledge service model, we address this issue in this article. For guaranteeing the inference efficiency of ontology knowledge, we propose a semiautomatic construction method for coal mine accident cases to construct ontology. Here, the reliability of ontology construction and the professionalism of domain knowledge provide a feasible approach to ontology learning using structured and semi-structured data sources. Furthermore, the weighted Word2vec and spectral clustering method are combined, an intelligent recommendation algorithm with accident ontology is accordingly presented, while presenting a local optimal distance calculation similarity strategy. This method can serve as an assistant to help users mine similar coal mine accident cases. Finally, the experimental results show that after comparing with some other popular methods, the proposed approach can achieve a satisfactory performance on the coal mine dataset with an accuracy of 99.47%, the precision of 98.92%, and the F1-score of 99.35.

关键词： domain knowledge intelligent recommendation ontology knowledge text similarity

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：