检索结果-内蒙古大学图书馆

SceneFormer: Neural Architecture Search of Transformers for Remote Sensing Scene Classification

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 2025年 63卷

作者： Tong, Lyuyang Liu, Jie Du, Bo Wuhan Univ Inst Artificial Intelligence Natl Engn Res Ctr Multimedia Software Sch Comp Sci Wuhan 430072 Peoples R China Wuhan Univ Hubei Key Lab Multimedia & Network Commun Engn Wuhan 430072 Peoples R China

Deep learning-based scene classification methods have long been a key research area in remote sensing imagery due to their wide-ranging applications. Recently, Transformer models have achieved significant progress in computer vision, making vision transformers (ViTs) a promising direction for scene classification. However, the spatial complexity of remote sensing imagery poses unique challenges for applying Transformers directly. Manually designing Transformers tailored for remote sensing scene classification is time-consuming under model parameter constraints and requires extensive domain expertise. To address this challenge, neural architecture search (NAS) methods provide an effective solution to construct optimal Transformer architectures for remote sensing scene classification automatically. In this work, we propose SceneFormer, an automated Transformer architecture search framework tailored for scene classification tasks. In SceneFormer, we construct a dedicated search space to search for the optimal Transformer. Moreover, we design a supernet training strategy to train numerous candidate architectures within the search space simultaneously. Furthermore, SceneFormer employs the evolutionary search to find the optimal Transformer architecture under specific resource constraints. Experiments on three high-spatial-resolution (HSR) datasets demonstrate the effectiveness of SceneFormer.

关键词： Transformers Computer architecture Remote sensing Scene classification Computer vision evolutionary computation Training computational modeling Neural architecture search Convolutional neural networks evolutionary algorithm neural architecture search (NAS) remote sensing scene classification vision transformer (ViT)

来源：评论

学校读者我要写书评

暂无评论

MODSC: Many-Objective-Optimization-Driven Data-Balancing Strategy in Cross-Architectural Malware Classification for Extreme IoT

引用

IEEE INTERNET OF THINGS JOURNAL 2024年第3期11卷 3702-3710页

作者： Cui, Zhihua Zhang, Zhigang Zhang, Zhixia Zhang, Wensheng Chen, Jinjun Taiyuan Univ Sci & Technol Shanxi Key Lab Big Data Anal & Parallel Comp Taiyuan 030024 Peoples R China Chinese Acad Sci Inst Automat Beijing 100045 Peoples R China Swinburne Univ Technol Dept Comp Technol Melbourne Vic 3122 Australia

Stable operation of IoT systems in extreme environments is critical for infrastructures, including energy, aerospace, and healthcare. But resource-constrained devices are vulnerable to malware exploiting known weaknesses due to complex operating conditions and low-frequency device maintenance, and systems are at serious risk of failure or information leakage. However, automated security defense based on machine learning faces the challenges of low cross-architectural generalization and multidimensional data imbalance. We propose a many-objective-optimization-driven data balancing strategy for cross-architectural malware classification (MODSC) to address the above issues. MODSC constructs the optimization problem model for data balancing strategy search based on data set information, which focuses on rebalancing the data space in different dimensions, including category distribution and architectural distribution. The many-objective evolutionary algorithm (MaOEA-PBF), guided by the performance balance function, is then designed to solve the above model. The MODSC framework was tested on a group of IoT malware data sets with stepped imbalance rates containing six common architectures. The experimental results show that MODSC can both steadily and effectively improve cross-architectural generalization while maintaining a high level of confidence compared to popular data processing methods.

关键词： Internet of Things Malware Computer architecture Optimization Training Performance evaluation evolutionary computation Cross-architecture extreme environments imbalanced learning malware detection many-objective optimization

来源：评论

学校读者我要写书评

暂无评论

Radioactivity estimation of radioactive hotspots using a Compton camera and derivation of dose rates in the surrounding environment

引用

APPLIED RADIATION AND ISOTOPES 2024年 212卷 111421页

作者： Sato, Yuki Japan Atom Energy Agcy Collaborat Labs Adv Decommissioning Sci 790-1 Motooka Aza Ohtsuka Tomioka Fukushima Japan

At the Fukushima Daiichi Nuclear Power Station, radiation sources released in the accident were deposited on various equipment and building structures. During decommissioning, it is crucial to understand the distribution of radiation sources and ambient dose equivalent rates to reduce worker exposure and implement detailed work planning. In this study, the author introduces a method for visualizing radiation sources, estimates their radioactivity using a Compton camera, and derives the dose rate around the radiation sources. In the demonstration test, the Compton camera was used to visualize radioactive hotspots caused by 137Cs radiation sources deposited in the outdoor environment and estimated the radioactivity. Furthermore, the dose rate around the hotspots was calculated from the estimated radioactivity, which confirmed that the calculated dose rate correlated with the dose rate measured using a survey meter. This approach is novel, where a series of analyses were conducted using the Compton camera to visualize radioactive hotspots, estimate the radioactivity, and derive the dose rate in the surrounding environment.

关键词： Gamma-ray imaging Compton camera Radioactivity estimation evolutionary computation Dose rate calculation

来源：评论

学校读者我要写书评

暂无评论

Exploring evolutionary Algorithms for Multi-Objective Optimization in Seismic Structural Design

引用

APPLIED SCIENCES-BASEL 2024年第21期14卷 9951页

作者： Korpeoglu, Seda Goktepe Yilmaz, Suleyman Mesut Yildiz Tech Univ Dept Math Engn TR-34220 Istanbul Turkiye

The seismic design of structures is an emerging practice in earthquake-resistant construction. Therefore, using energy-dissipation devices and optimizing these devices for various purposes are important. evolutionary computation, nature-inspired, and meta-heuristic algorithms have been studied more in recent years for the optimization of these devices. In this study, the development of evolutionary algorithms for seismic design in the context of multi-objective optimization is examined through bibliometric analysis. In particular, evolutionary algorithms such as genetic algorithms and particle swarm optimization are used to optimize the performance of structures to meet seismic loads. While genetic algorithms are used to improve both the cost and seismic performance of the structure, particle swarm optimization is used to optimize the vibration and displacement performance of structures. In this study, a bibliometric analysis of 661 publications is performed on the Web of Science and Scopus databases and on how the research in this field has developed since 1986. The R-studio program with the biblioshiny package is used for the analyses. The increase in studies on the optimization of energy dissipation devices in recent years reveals the effectiveness of evolutionary algorithms in this field.

关键词： multi-objective optimization seismic design evolutionary computation nature-inspired algorithms metaheuristics bibliometric analysis

来源：评论

学校读者我要写书评

暂无评论

Hierarchically Composing Level Generators for the Creation of Complex Structures

引用

IEEE TRANSACTIONS ON GAMES 2024年第2期16卷 459-469页

作者： Beukman, Michael Fokam, Manuel Kruger, Marcel Axelrod, Guy Nasir, Muhammad Ingram, Branden Rosman, Benjamin James, Steven Univ Witwatersrand Sch Comp Sci & Appl Math Johannesburg South Africa

Procedural content generation (PCG) is a growing field, with numerous applications in the video game industry and great potential to help create better games at a fraction of the cost of manual creation. However, much of the work in PCG is focused on generating relatively straightforward levels in simple games, as it is challenging to design an optimizable objective function for complex settings. This limits the applicability of PCG to more complex and modern titles, hindering its adoption in the industry. Our work aims to address this limitation by introducing a compositional level generation method that recursively composes simple low-level generators to construct large and complex creations. This approach allows for easily-optimizable objectives and the ability to design a complex structure in an interpretable way by referencing lower-level components. We empirically demonstrate that our method outperforms a noncompositional baseline by more accurately satisfying a designer's functional requirements in several tasks. Finally, we provide a qualitative showcase (in Minecraft) illustrating the large and complex, but still coherent, structures that were generated using simple base generators.

关键词： Games Generators Task analysis Genetic algorithms Statistics Sociology Search problems evolutionary computation genetic algorithms procedural content generation

来源：评论

学校读者我要写书评

暂无评论

Privacy-preserving data publishing: an information-driven distributed genetic algorithm

引用

WORLD WIDE WEB-INTERNET AND WEB INFORMATION SYSTEMS 2024年第1期27卷 1-1页

作者： Ge, Yong-Feng Wang, Hua Cao, Jinli Zhang, Yanchun Jiang, Xiaohong Victoria Univ Melbourne Australia La Trobe Univ Melbourne Australia Zhejiang Normal Univ Sch Comp Sci & Technol Jinhua Zhejiang Peoples R China Peng Cheng Lab Dept New Networks Shenzhen Guangdong Peoples R China Victoria Univ Inst Sustainable Ind & Liveable Cities Melbourne Vic Australia Future Univ Hakodate Hakodate Japan

The privacy-preserving data publishing (PPDP) problem has gained substantial attention from research communities, industries, and governments due to the increasing requirements for data publishing and concerns about data privacy. However, achieving a balance between preserving privacy and maintaining data quality remains a challenging task in PPDP. This paper presents an information-driven distributed genetic algorithm (ID-DGA) that aims to achieve optimal anonymization through attribute generalization and record suppression. The proposed algorithm incorporates various components, including an information-driven crossover operator, an information-driven mutation operator, an information-driven improvement operator, and a two-dimensional selection operator. Furthermore, a distributed population model is utilized to improve population diversity while reducing the running time. Experimental results confirm the superiority of ID-DGA in terms of solution accuracy, convergence speed, and the effectiveness of all the proposed components.

关键词： evolutionary computation Data privacy and utility Data publishing Distributed algorithm

来源：评论

学校读者我要写书评

暂无评论

Identifying Pareto Fronts Reliably Using a Multistage Reference-Vector-Based Framework

引用

IEEE TRANSACTIONS ON evolutionary computation 2024年第1期28卷 252-266页

作者： Deb, Kalyanmoy Lopes, Claudio Lucio do Val Martins, Flavio Vinicius Cruzeiro Wanner, Elizabeth Fialho Michigan State Univ Dept Elect & Comp Engn E Lansing MI 48824 USA CEFET MG Postgrad Program Math & Computat Modeling BR-30421169 Belo Horizonte Brazil Ctr Fed Educ Tecnol Minas Gerais Comp Dept Belo Horizonte Brazil

evolutionary multiobjective and many-objective optimization (EMO and EMaO) algorithms are increasingly used to identify the true shape and location of the Pareto-optimal front using a few representative well-converged and well-distributed solutions. The reason for their popularity is due to their ability to provide a better understanding of objective relationships for optimal solutions, and also to facilitate the choice of a preferred solution using an interactive or post-optimal multicriterion decision analysis. However, since EMO and EMaO algorithms are stochastic, a single application may not provide a true representative set with a desired number of Pareto solutions reliably in repetitive runs and importantly with a well-distributed set of solutions. In this article, we propose a multistage framework involving reference-vector-based evolutionary multi- and many-objective algorithms (MuSt-EMO and MuSt-EMaO) that attempts to recursively rectify shortcomings of previous stages by careful executions of subsequent stages so that a prescribed number of well-distributed and well-converged solutions are achieved at the end. The proposed multistage approach is implemented to a number of popular reference vector-based EMO/EMaO algorithms and is applied on various multi- and many-objective test and real-world problems.

关键词： Convergence Statistics Sociology Optimization Reliability Heuristic algorithms evolutionary computation evolutionary algorithms (EAs) gap-filling method multiobjective optimization pareto front

来源：评论

学校读者我要写书评

暂无评论

An Objective Space Constraint-Based evolutionary Method for High-Dimensional Feature Selection

引用

IEEE computationAL INTELLIGENCE MAGAZINE 2024年第2期19卷 113-128页

作者： Cheng, Fan Zhang, Rui Huang, Zhengfeng Qiu, Jianfeng Xia, Mingming Zhang, Lei Anhui Univ Hefei Peoples R China Hefei Univ Technol Hefei Peoples R China

evolutionary algorithms (EAs) have shown their competitiveness in solving the problem of feature selection. However, limited by their encoding scheme, most of them face the challenge of "curse of dimensionality". To address the issue, in this paper, an objective space constraint-based evolutionary algorithm, named OSC-EA, is proposed for high-dimensional feature selection (HDFS). Although the decision space of EAs for HDFS is very huge, its objective space is the same as that of the low-dimensional feature selection. Based on this fact, in the proposed OSC-EA, the HDFS is firstly modeled as a constrained problem, where a constraint of the objective space is introduced and used to partition the whole objective space into the "feasible region" and the "infeasible region". To handle the constrained problem, a two-stage epsilon constraint-based evolutionary scheme is designed. In the first stage, the value of epsilon is set to be very small, which ensures that the search concentrates on the "feasible region", and the latent high-quality feature subsets can be found quickly. Then, in the second stage, the value of epsilon increases gradually, so that more solutions in the "infeasible region" are considered. Until the end of the scheme, epsilon -> infinity;all the solutions in the objective space are considered. By using the search in the second stage, the quality of the obtained feature subsets is further improved. The empirical results on different high-dimensional datasets demonstrate the effectiveness and efficiency of the proposed OSC-EA.

关键词： evolutionary computation Encoding Feature extraction Constraint handling

来源：评论

学校读者我要写书评

暂无评论

A hierarchical clustering algorithm for addressing multi-modal multi-objective optimization problems

引用

EXPERT SYSTEMS WITH APPLICATIONS 2025年 264卷

作者： Gu, Qinghua Niu, Yiwen Hui, Zegang Wang, Qian Xiong, Naixue Xian Univ Architecture & Technol Sch Resources Engn Xian 710055 Shaanxi Peoples R China Xian Univ Architecture & Technol Xian Key Lab Intelligent Ind Percept Calculat & De Xian 710055 Peoples R China Xian Univ Architecture & Technol Sch Management Xian 710055 Shaanxi Peoples R China James Madison Univ Integrated Sci & Technol Harrisonburg VA 22807 USA Ross State Univ Dept Comp Math & Phys Sci Alpine TX 79830 USA

Many evolutionary multi-modal multi-objective algorithms (MMEAs) have been proposed to solve multi-modal multi-objective optimization problems (MMOPs). Unfortunately, the environmental selection process causes many algorithms to place too much emphasis on solution variety in the decision space, which leads to solutions with low convergence quality. As a result, not only are all local Pareto fronts reversed, but objective values are also far lower than the global Pareto Fronts. To tackle these tasks, this paper proposes a hierarchical clustering-based MMOEA_DC_HR model that uses decision space clustering methods to group neighborhood solutions into several local clusters, preserving local Pareto Sets. And secondary clustering is performed in the objective space to select temporary populations from these local clusters to maintain the diversity of the objective space. Additionally, a hierarchical ranking method is introduced to update the convergence archive, aiding in maintaining the convergence of the algorithm and controlling the quality of the Pareto Front. The test results show that this novel algorithm exhibits competitive performance in solving selected benchmark problems when compared to other cutting-edge MMEAs.

关键词： Multi-modal multi-objective optimization evolutionary computation Hierarchical clustering Local convergence

来源：评论

学校读者我要写书评

暂无评论

Hybrid Ant Lion Mutated Ant Colony Optimizer Technique With Particle Swarm Optimization for Leukemia Prediction Using Microarray Gene Data

引用

IEEE ACCESS 2024年 12卷 10910-10919页

作者： Mahesh, T. R. Santhakumar, D. Balajee, A. Shreenidhi, H. S. Kumar, V. Vinoth Rajkumar Annand, Jonnakuti JAIN Fac Engn & Technol Dept Comp Sci & Engn Bengaluru 562112 India SIMATS Saveetha Sch Engn Chennai 602107 India Vellore Inst Technol VIT Sch Comp Sci Engn & Informat Syst SCORE Vellore 632014 Tamil Nadu India Arba Minch Univ Dept Electromech Engn Sawla Campus Arba Minch 4400 Ethiopia

Leukemia refers to a type of blood malignancy that develops due to certain hematological disorders. Identifying leukemia at its earlier stages through clinical operations are highly complicated task with invasive methods. Gene expression data could be collected and computational methods could be adopted which could lead to better prediction of leukemia that leads to prevention at its earlier stages. Today, feature selection has become an important step in pre-processing that helps bring improvement to the classification system and its performance that is done by choosing optimal feature subsets by means of reducing or eliminating redundant or irrelevant features. Particle Swarm Optimization (PSO) is a popular algorithm wherein certain solutions that are generated randomly move within the search space to obtain optimal solutions. Another relatively new and evolutionary method computation is the Ant Lion Optimization (ALO) algorithm that has lower computation cost compared to the other techniques. In this work, a new technique known as the Hybrid Ant Lion Mutated Ant Colony Optimize along with Particle Swarm Optimization (PSO) was proposed for the prediction of leukaemia with the microarray gene data. The proposed model that is used for identifying the optimal set of features from which the classification has been done using the Support Vector Machine (SVM) has produced a significant prediction accuracy of 87.88%.

关键词： Feature extraction Optimization Gene expression Classification algorithms Support vector machines Search problems Particle swarm optimization Leukemia gene expression data feature selection ant lion optimization (ALO) algorithm evolutionary computation

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：