检索结果-内蒙古大学图书馆

An Improved MPCA Algorithm with Weight Matrix Based on Many-Objective Optimization 1

17th International Conference on Bio-Inspired Computing: Theories and Applications, BIC-TA 2022

作者： Huang, Jianrou Zhang, Jingbo Wang, Qian Cai, Xingjuan Complex System and Computational Intelligent Laboratory Taiyuan University of Science and Technology Taiyuan030024 China

ISBN: (数字)9789819915491

ISBN: (纸本)9789819915484

Data compression and dimensionality reduction play an essential role in machine learning. In recent years, due to the relatively high dimensionality of data, dimensionality reduction based on vectors cannot be well processed and will destroy the structure of high-dimensional data. Therefore, research on dimensionality reduction based on tensors is increasing daily. This paper mainly studies the extended multi-linear principal component analysis (MPCA) of principal component analysis (PCA) in the tensor space. This paper proposes a MPCA dimension reduction algorithm based on the weight matrix, followed by the construction of a multi-objective model. An IP-NSGA-III is proposed to solve the weight matrix. Experimental results of hyperspectral image classification show that, compared with different algorithms, the MPCA algorithm based on the weight matrix proposed in this paper has a better dimension-reduction effect. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Principal component analysis

来源：评论

学校读者我要写书评

暂无评论

Research on Explainable Recommendation Model Based on Knowledge Graph 19

Research on Explainable Recommendation Model Based on Knowle...

引用

19th IEEE International Conference on Networking, Sensing and Control, ICNSC 2022

作者： Xie, Lijie Jin, Yaqing Cui, Zhihua Wang, Lifang Taiyuan University Science and Technology Complex System and Computational Intelligence Laboratory Taiyuan China

ISBN: (纸本)9781665472432

In order to further obtain the best explanation paths and recommendation items that satisfy the preferences of users accurately and efficiently, a many-objective explainable recommendation model GrEA-ERS based on preferences of users is proposed, which designs pruning strategies to delete unsatisfied paths in knowledge graph based on on preferences of users, so as to reduce the search paths. Combined with the many-objective optimization algorithm GrEA, the paths in knowledge graph are used as decision variables, at which different explained paths correspond to different recommended items, and accuracy, diversity, novelty and explainability are optimized to obtain explained paths and their corresponding recommended items. Through extensive experiments, the model GrEA-ERS always outperforms other algorithms in terms of accuracy, novelty, diversity, and explainability. © 2022 IEEE.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

A Many-Objective Anomaly Detection Model for Vehicle Network Based on Federated Learning and Differential Privacy Protection 12th

A Many-Objective Anomaly Detection Model for Vehicle Network...

引用

12th International Symposium on Intelligence Computation and Applications, ISICA 2021

作者： Fan, Tian Zhang, Zhixia Lan, Yang Cui, Zhihua Complex System and Computational Intelligence Laboratory Taiyuan University of Science and Technology Taiyuan China

ISBN: (纸本)9789811941085

With the rapid development of Internet of things, vehicle network has become the focus of smart city construction. However, due to the lack of encryption and authentication mechanism, the vehicle network is vulnerable to malicious attacks and intrusion, which poses a threat to the driver’s life safety and data privacy. To solve this problem, this paper proposes a method combining federal learning and differential privacy protection to protect the privacy of drivers. A many-objective anomaly detection model based on Federated learning is constructed. The model optimizes the accuracy, loss, privacy protection degree and communication cost in the federated training model at the same time. The model can obtain better accuracy and privacy protection with less loss and communication cost. The model is solved by many-objective evolutionary algorithm, and some solutions are selected as training parameters, and satisfactory results are obtained. The effectiveness of the proposed method is verified. © 2022, Springer Nature Singapore Pte Ltd.

关键词： Vehicles

来源：评论

学校读者我要写书评

暂无评论

Parallel Reinforcement Learning and Gaussian Process Regression for Improved Physics-Based Nasal Surgery Planning 15th

Parallel Reinforcement Learning and Gaussian Process Regres...

引用

15th International Conference on Parallel Processing and Applied Mathematics, PPAM 2024

作者： Rüttgers, Mario Hübenthal, Fabian Tsubokura, Makoto Lintermann, Andreas Jülich Supercomputing Centre Forschungszentrum Jülich Jülich52428 Germany RWTH Aachen University Aachen52062 Germany Department of Computational Science Graduate School of System Informatics Kobe University Kobe657-8501 Japan Complex Phenomena Unified Simulation Research Team RIKEN Center for Computational Science Kobe650-0047 Japan

ISBN: (纸本)9783031857027

Septoplasty and turbinectomy are among the most frequent but also most debated interventions in the field of rhinology. A previously developed tool enhances surgery planning by physical aspects of respiration, i.e., for the first time a reinforcement learning (RL) algorithm is combined with large-scale computational fluid dynamics (CFD) simulations to plan anti-obstructive surgery. In the current study, an improvement of the tool’s predictive capabilities is investigated for the aforementioned types of surgeries by considering two approaches: (i) training of parallel environments is executed on multiple ranks and the agents of each environment share their experience in a pre-defined interval and (ii) for some of the state-reward combinations the CFD solver is replaced by a Gaussian process regression (GPR) model for an improved computational efficiency. It is found that employing a parallel RL algorithm improves the reliability of the surgery planning tool in finding the global optimum. However, parallel training leads to a larger number of state-reward combinations that need to be computed by the CFD solver. This overhead is compensated by replacing some of the computations with the GPR algorithm, i.e., around 6% of the computations can be saved without significantly degrading the predictions’ accuracy. Nevertheless, increasing the number of state-reward combinations predicted by the GPR algorithm only works to a certain extent, since this also leads to larger errors. © The Author(s) 2025.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

A new class of generalized quasi-variational inequalities with applications to Oseen problems under nonsmooth boundary conditions

引用

Science China Mathematics 2024年第2期67卷 315-338页

作者： Shengda Zeng Akhtar A.Khan Stanislaw Migórski Guangxi Colleges and Universities Key Laboratory of Complex System Optimization and Big Data Processing Yulin Normal UniversityYulin 537000China Faculty of Mathematics and Computer Science Jagiellonian University in Krak´owKrak´ow 30348Poland Center for Applied and Computational Mathematics School of Mathematical SciencesRochester Institute of TechnologyRochesterNY 14623USA College of Sciences Beibu Gulf UniversityQinzhou 535000China

In this paper, we study a generalized quasi-variational inequality (GQVI for short) with twomultivalued operators and two bifunctions in a Banach space setting. A coupling of the Tychonov fixedpoint principle and the Katutani-Ky Fan theorem for multivalued maps is employed to prove a new existencetheorem for the GQVI. We also study a nonlinear optimal control problem driven by the GQVI and givesufficient conditions ensuring the existence of an optimal control. Finally, we illustrate the applicability of thetheoretical results in the study of a complicated Oseen problem for non-Newtonian fluids with a nonmonotone andmultivalued slip boundary condition (i.e., a generalized friction constitutive law), a generalized leak boundarycondition, a unilateral contact condition of Signorini’s type and an implicit obstacle effect, in which themultivalued slip boundary condition is described by the generalized Clarke subgradient, and the leak boundarycondition is formulated by the convex subdifferential operator for a convex superpotential.

关键词： generalized quasi-variational inequality existence theorem optimal control Kakutani-Ky Fan theorem Oseen problem non-Newtonian fluid nonmonotone slip boundary condition

来源：评论

学校读者我要写书评

暂无评论

Parallels between drought and flooding: An integrated framework for plant eco-physiological responses to water stress

引用

Plant-Environment Interactions 2023年第4期4卷 175-187页

作者： Chen, Siluo ten Tusscher, Kirsten H. W. J. Sasidharan, Rashmi Dekker, Stefan C. de Boer, Hugo J. Computational Developmental Biology Department of Biology Utrecht University Utrecht Netherlands Centre for Complex System Studies Utrecht University Utrecht Netherlands Plant Stress Resilience Institute of Environmental Biology Utrecht University Utrecht Netherlands Environmental Sciences Copernicus Institute of Sustainable Development Utrecht University Utrecht Netherlands

Drought and flooding occur at opposite ends of the soil moisture spectrum yet their resulting stress responses in plants share many similarities. Drought limits root water uptake to which plants respond with stomatal closure and reduced leaf gas exchange. Flooding limits root metabolism due to soil oxygen deficiency, which also limits root water uptake and leaf gas exchange. As drought and flooding can occur consecutively in the same system and resulting plant stress responses share similar mechanisms, a single theoretical framework that integrates plant responses over a continuum of soil water conditions from drought to flooding is attractive. Based on a review of recent literature, we integrated the main plant eco-physiological mechanisms in a single theoretical framework with a focus on plant water transport, plant oxygen dynamics, and leaf gas exchange. We used theory from the soil–plant–atmosphere continuum modeling as “backbone” for our framework, and subsequently incorporated interactions between processes that regulate plant water and oxygen status, abscisic acid and ethylene levels, and the resulting acclimation strategies in response to drought, waterlogging, and complete submergence. Our theoretical framework provides a basis for the development of mathematical models to describe plant responses to the soil moisture continuum from drought to flooding. © 2023 The Authors. Plant-Environment Interactions published by New Phytologist Foundation and John Wiley & Sons Ltd.

关键词： drought ethylene flooding modeling oxygen deficit soil-plant-atmosphere continuum water stress

来源：评论

学校读者我要写书评

暂无评论

Research on Explainable Recommendation Model Based on Knowledge Graph

Research on Explainable Recommendation Model Based on Knowle...

引用

IEEE International Conference on Networking, Sensing and Control

作者： Lijie Xie Yaqing Jin Zhihua Cui Lifang Wang Complex System and Computational Intelligence Laboratory Taiyuan University Science and Technology Taiyuan Peoples R China

ISBN: (纸本)9781665472449

关键词： Knowledge engineering Semantics Sensors Optimization

来源：评论

学校读者我要写书评

暂无评论

Optimal hydropower station dispatch using quantum social spider optimization algorithm

Optimal hydropower station dispatch using quantum social spi...

引用

作者： Zhou, Guo Zhao, Ruxin Luo, Qifang Zhou, Yongquan Department of Science and Technology Teaching China University of Political Science and Law Beijing China Key Laboratories of Guangxi High Schools Complex System and Computational Intelligence Nanning China School of Information Science and Engineering Guangxi University for Nationalities Nanning China

In this article, a new quantum social spider optimization (QSSO) algorithm is proposed. In the QSSO algorithm, we introduce an encoding approach based on bits described on social spider optimization (SSO) and serves as the evolution method of the population space. For the encoding of individuals, the probability amplitude expression of quantum bit is applied to describe the position of individuals, by which one individual's position can be expressed as the superposition of multistates. In such a way, the population diversity and the global searching capability of the SSO algorithm are enhanced. The QSSO algorithm was used to optimize the hydropower station dispatch, and the calculation results show that QSSO algorithm has fast convergence, small number of tuning parameters, high calculation accuracy, stability, simple, and is easy to be implemented with strong global search capability. © 2020 John Wiley & Sons, Ltd.

关键词： Optimization

来源：评论

学校读者我要写书评

暂无评论

Demand estimation of water resources via bat algorithm

引用

International Journal of Wireless and Mobile Computing 2020年第1期18卷 16-21页

作者： Pei, Xiangdong Sun, Youqiang Ren, Yeqing Taiyuan Comprehensive Senior High School Taiyuan China Complex System and Computational Intelligence Laboratory Taiyuan University of Science and Technology Taiyuan China

In the process of urban water resources planning, the demand estimation of urban water consumption is one of the important basic contents. In this paper, a hybrid model of linear estimation model and an exponential estimation model are proposed to forecast the water consumption. The bionic intelligent algorithms are widely used in industrial engineering, so, we use intelligent algorithms to solve the proposed model including Bat Algorithm (BA) and modified Bat Algorithm (FTBA). FTBA improves the global search capability, and the improvements increase the probability of solving the optimal value. In the simulation experiments, we use the data from Nanchang city during 2003 to 2015. The data from 2003 to 2012 are used to find the optimal weights, and the remaining data (2013–2015) are used to test the models. Simulation results show that the modified BA (FTBA) is superior to the standard algorithm and achieves higher accuracy in prediction. Copyright © 2020 Inderscience Enterprises Ltd.

关键词： Water resources

来源：评论

学校读者我要写书评

暂无评论

A Many-Objective Algorithm with Threshold Elite Selection Strategy 11th

A Many-Objective Algorithm with Threshold Elite Selection St...

引用

11th International Symposium on Intelligence Computation and Applications, ISICA 2019

作者： Geng, Shaojin Wu, Di Wang, Penghong Cai, Xingjuan Complex System and Computational Intelligence Laboratory Taiyuan University of Science and Technology Taiyuan China

ISBN: (纸本)9789811555763

The study of many-objective evolutionary algorithm (MaOEA) has become particularly important, especially with the increasing complex engineering optimization problems. Considering that the convergence and diversity of the population are two important indicators to measure the performance of the algorithm, a many-objective evolutionary algorithm with threshold elite selection strategy (MaOEA-TES) are proposed in this paper. The algorithm adopts the balanceable fitness estimation strategy and the reference-point based non-dominated sorting strategy to balance the convergence and diversity of the solution. An adaptive penalty distance boundary intersection strategy is designed to dynamically adjust the impact of convergence and diversity on the algorithm. In addition, a dynamic threshold selection strategy is proposed to ensure that the algorithm emphasizes diversity at an early stage, emphasizes convergence at a later stage, and ensures that the result is closer to the real non-dominant front. The DTLZ test suite is used to evaluate the performance of MaOEA-TES. The experimental results show that the MaOEA-TES has the best performance comparing with three other state-of-the-art algorithms on many-objective optimization. © 2020, Springer Nature Singapore Pte Ltd.

关键词： Evolutionary algorithms

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：