检索结果-内蒙古大学图书馆

13th IEEE Congress on Evolutionary Computation, CEC 2024

作者： Kobayashi, Kosei Aritsugi, Masayoshi González, Pedro Henrique Mendonça, Israel Graduate School of Science and Technology Kumamoto University Kumamoto Japan Faculty of Advanced Science and Technology Kumamoto University Kumamoto Japan System Engineering and Computer Science Program Federal University of Rio de Janeiro Rio de Janeiro Brazil

ISBN: (纸本)9798350308365

Hyperparameter optimization (HPO) is paragon to maximize performance when designing machine learning models. Among different HPO methods, Genetic Algorithm (GA) based optimization is considered effective because it allows a wide and diverse range of solutions to be explored. However, GA's exploratory nature makes this type of algorithm to evaluate many solutions that do not improve the overall performance. This is specially costly when the objective function to be evaluated is time-consuming, like in the HPO field. In this paper, we propose an efficient hybrid algorithm that is able to reduce computational cost by combining deep reinforcement learning with the Biased Random Key Genetic Algorithm (BRKGA), a variant of genetic algorithms. Our reinforcement learning agent has a decision-making role during the population's fitness calculation, in which it filters out chromosomes that would not improve the overall fitness of the population. The agent uses small amounts of pre-trained data to identify trends in potentially good solutions, and carry out its decision process. We conduct experiments on eight different datasets to assess the effectiveness of the proposed method, and the results show that the proposed method can significantly reduce the computation time of hyperparameter search using BRKGA (up to 44% reduction in computational time) without compromising the quality of the solution (no statistically difference in results). © 2024 IEEE.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

An Improved Transfer Learning-Based Model with Data Augmentation for Brain Tumor Detection 6

An Improved Transfer Learning-Based Model with Data Augmenta...

引用

6th International Conference on Artificial Intelligence in Information and Communication, ICAIIC 2024

作者： Prayogo, Rizal Dwi Hamid, Nur Nambo, Hidetaka Graduate School of Natural Science and Technology Kanazawa University Division of Electrical Engineering and Computer Science Kanazawa Japan Institute of Philosophy in Interdisciplinary Sciences Kanazawa University Faculty of Transdisciplinary Sciences Kanazawa Japan

ISBN: (纸本)9798350344349

Current advances in deep learning have brought various breakthroughs in processing medical data. However, dealing with a limited number of medical datasets remains a challenge in deep learning and often leads to overfitting. To solve this research gap, here we show a new approach to improve the performance of a transfer learning-based model for brain tumor detection from 253 brain magnetic resonance imaging (MRI) sample images. The concept of transfer learning has been applied using a pretrained Inception V3combined with data augmentation. Modified layers using dropout and regularization have been additionally utilized to deal with overfitting. The proposed method shows an increase in accuracy of 6.430%, a precision of 5.531 %, a recall of 10.545%, and an F1-score of 8.040% compared to the baseline method. We show that our proposed method has been able to effectively enhance performance and reduce overfitting, even with a small number of datasets. Moreover, our proposed method outperforms state-of-The-Art brain tumor detection. © 2024 IEEE.

关键词： Tumors

来源：评论

学校读者我要写书评

暂无评论

Prediction Of Power Demand And Factor Identification Using Sparse Modeling AI With Ensemble Learning For Improvement Of Accuracy

Prediction Of Power Demand And Factor Identification Using S...

引用

International Symposium on Soft Computing and Intelligent systems (SCIS)

作者： Tatsuma Hachiya Keiji Kamei Graduate School of Life Science and Systems Engineering Kyushu Institute of Technology Fukuoka Japam Faculty of Engineering Nishinippon Institute of Technology Fukuoka Japan

ISBN: (数字)9798350373332

ISBN: (纸本)9798350373349

We have been using sparse modeling AI to identify factors and predict in power demand. In this research, we applied ensemble learning to further improve the accuracy by integrating the results of multiple models. Experimental results showed that sparse modeling AI identified the demand factors, and ensemble learning achieved an average absolute prediction error of 7.5%.

关键词： Power demand Accuracy Atmospheric modeling Electricity Predictive models Ensemble learning Artificial intelligence Tuning Meteorology

来源：评论

学校读者我要写书评

暂无评论

Efficient Node Reduction Heuristic for GNN-Based Traffic Speed Forecasting 11

Efficient Node Reduction Heuristic for GNN-Based Traffic Spe...

引用

11th IEEE/ACM International Conference on Big Data Computing, Applications and Technologies, BDCAT 2024

作者： Inokuchi, Yuto Gonzalez, Pedro Henrique Aritsugi, Masayoshi Mendonca, Israel Kumamoto University Graduate School of Science and Technology Kumamoto Japan Federal University of Rio de Janeiro System Engineering and Computer Science Program Rio de Janeiro Brazil Kumamoto University Faculty of Advanced Science and Technology Kumamoto Japan

ISBN: (纸本)9798350367300

In this paper, we aim to reduce the number of nodes from Graph Neural Networks (GNNs), thereby simplifying models and reducing computational costs. GNNs are highly effective for various tasks, such as prediction, classification, and clustering, due to their ability to learn node and edge attributes and relationships, and they have been utilized for intelligent transportation systems recently by converting sensor networks into graph structures. Deep spatio-temporal neural networks, including Spatio-Temporal Graph Convolutional Networks (STGCNs), capture spatial and temporal dependencies, making them suitable for traffic speed forecasting, traffic demand prediction, and travel time estimation. Despite their success, GNNs face challenges in industrial applications due to significant memory usage and time consumption. In this paper, we propose a new approach to node reduction that outperforms existing methods in computational efficiency. Our experiments on two real-world traffic datasets demonstrate that using the heuristic and edge information to reduce nodes can cut computation time of optimization up to 95% and, by eliminating noise, can even enhance prediction accuracy. © 2024 IEEE.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Dynamic S-Box Generation Using Novel Chaotic Map with Nonlinearity Tweaking

引用

Computers, Materials & Continua 2023年第5期75卷 3011-3026页

作者： Amjad Hussain Zahid Muhammad Junaid Arshad Musheer Ahmad Naglaa F.Soliman Walid El-Shafai School of Systems and Technology University of Management and TechnologyLahore54700Pakistan Department of Computer Science University of Engineering and TechnologyLahore54700Pakistan Department of Computer Engineering Jamia Millia IslamiaNew Delhi110025India Department of Information Technology College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityP.O.Box 84428Riyadh11671Saudi Arabia Security Engineering Lab Computer Science DepartmentPrince Sultan UniversityRiyadh11586Saudi Arabia Department of Electronics and Electrical Communications Engineering Faculty of Electronic EngineeringMenoufia UniversityMenouf32952Egypt

A substitution box(S-Box)is a crucial component of contemporary cryptosystems that provide data protection in block *** the moment,chaotic maps are being created and extensively used to generate these SBoxes as a chaotic map assists in providing disorder and resistance to combat cryptanalytical *** this paper,the construction of a dynamic S-Box using a cipher key is proposed using a novel chaotic map and an innovative tweaking *** projected chaotic map and the proposed tweak approach are presented for the first time and the use of parameters in their workingmakes both of these dynamic in *** tweak approach employs cubic polynomials while permuting the values of an initial S-Box to enhance its cryptographic *** of the parameters are provided using the cipher key and a small variation in values of these parameters results in a completely different unique *** analysis and exploration confirmed that the projected chaoticmap exhibits a significant amount of chaotic *** security assessment in terms of bijectivity,nonlinearity,bits independence,strict avalanche,linear approximation probability,and differential probability criteria are utilized to critically investigate the effectiveness of the proposed S-Box against several *** proposed S-Box’s cryptographic performance is comparable to those of recently projected S-Boxes for its adaption in real-world security *** comparative scrutiny pacifies the genuine potential of the proposed S-Box in terms of its applicability for data security.

关键词： Substitution-box chaotic map data security tweaking

来源：评论

学校读者我要写书评

暂无评论

Improvement in Properties of Martensitic Stainless Steel SUS420J2 Fabricated by LPBF Method Through Heat Treatment

引用

Zairyo/Journal of the Society of Materials science, Japan 2025年第4期74卷 259-266页

作者： Taka, Takeaki Takubo, Ryosuke Takesue, Shogo Morita, Tatsuro Graduate School of Science and Technology Kyoto Institute of Technology Goshokaido-cho Matsugasaki Sakyo-ku Kyoto606-8585 Japan TOWA Corporation Kamichoshi-cho Kamitoba Minami-ku Kyoto601-8105 Japan Faculty of Mechanical Engineering Kyoto Institute of Technology Goshokaido-cho Matsugasaki Sakyo-ku Kyoto606-8585 Japan

In this study, martensitic stainless steel SUS420J2 (AM material) was fabricated by laser powder bed fusion (LPBF), one of the additive manufacturing (AM) methods, using the laser scanning pattern that was selected in our previous study as a suitable one to reduce residual stress. AM material was heat-treated under two conditions, that is, conventional quenching and tempering (H1 treatment) and only tempering at low temperature for a long time (H2 treatment), and their effects were systematically investigated. The results showed that the amount of the retained γ phase in the microstructure of AM material significantly reduced with the H1 treatment and the microstructure became almost the same to that of the H1-treated wrought material (W material). Hardness, static strength and wear resistance of the H1-treated AM material were comparable to those of the H1-treated W material. Moreover, the H1 treatment gave the effect to markedly reduce the directional dependence and the absolute value of residual stress generated in AM material. On the other hand, the microstructure of AM material was unchanged by the H2 treatment. Further, its hardness and static strength were lower than those of the untreated AM material, besides the directional dependence and the absolute value of residual stress remained almost unchanged. The fatigue strength of W material was greatly improved by the H1 treatment, but the improvement of the fatigue strength of the H1-treated AM material was smaller than that of the H1-treated W material because the fatigue strength of AM material was significantly affected by molding defects, and the fatigue strength of the H2-treated AM material was not improved. © 2025 The Society of Materials science, Japan.

关键词： Tempering

来源：评论

学校读者我要写书评

暂无评论

CGTI-Net: Deep-Learning-Based Object Detection Network for High-Resolution Aerial Images

CGTI-Net: Deep-Learning-Based Object Detection Network for H...

引用

2023 IEEE International Conference on Internet of Things and Intelligence systems, IoTaIS 2023

作者： Horikawa, Yusei Mochizuki, Seiji Imamura, Kousuke Matsumura, Tetsuya Graduate School of Engineering and Technology Nihon University Fukushima Japan Osaka Sangyo University Faculty of Engineering Osaka Japan College of Science and Engineering Kanazawa University Ishikawa Japan College of Engineering Nihon University Fukushima Japan

ISBN: (纸本)9798350319040

We propose the Cluster-Guided Two-Stage Inference Network (CGTI-Net), a deep-learning-based object detection network for high-resolution aerial images. The network consists of an optimized two-stage inference structure, and a normalized clustering unit (NCU) to estimate small object dense areas. The two-stage inference structure is classified into a global detection unit (GDU) and a local detection unit (LDU). They are designed to detect large and small objects respectively. CGTI-Net improves the inference performance of small objects by estimating the location of small objects as clusters, thereby guiding the locations that perform fine inference. To estimate the clusters, CGTI-Net performs normalization and optimization of the vectors obtained through the GDU, after that it classifies them using a clustering algorithm. The vector optimization in the NCU focuses on the tendency that object vectors are correlated in terms of location and confidence score, in addition to vector components such as width and height, which determine their size. Then, the distance in vector space is biased according to the strength of the correlation coefficient of each vector component with respect to object size. Through optimization, the smaller the object size, the closer the distance in the vector space and the more likely it is to be classified into a cluster. CGTI-Net with Ward's method achieved the highest scores among the tested methods in terms of the number of detections, mean average precision, and small object detection. © 2023 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

WOOD-ONLY TIMBER FRAME JOINTS INSPIRED BY TREES’ BRANCH MODEL, PART1: STRUCTURAL SYSTEM AND BENDING PERFORMANCE

引用

AIJ Journal of technology and Design 2023年第73期29卷 1296-1301页

作者： Hawasly, Firas Matsumoto, Naoyuki Koshihara, Mikio Adachi, Koji Akita Prefectural Univ. Institute of Wood Technology Japan Tohoku Univ. Graduate School of Engineering Dept. of Architecture and Building Science Japan The Univ. of Tokyo Institute of Industrial Science Dept. of Human and Social Systems Japan

This paper aims to evaluate the design and bending performance of a wood-only timber frame joint inspired by the growth model in trees. The stem branch junction’s features were explored and approached in a simplified application of glued-laminated column-beam members joined in a conical mortise-tenon with bent laminated veneer plies flowing seamlessly across their interface. The joint was manufactured in Japanese Cedar with Beech bent plies, tested, and compared to glued-in rod (GIR) and lag screw bolt (LSB) joints;it showed comparable efficiency in moment transfer with reasonable failure patterns compared to conventional joint systems with added-steel accessories. © 2023 Architectural Institute of Japan. All rights reserved.

关键词： Laminating

来源：评论

学校读者我要写书评

暂无评论

Smart Data Augmentation for Multiclass Brain Tumor Detection with Imbalanced Data 8

Smart Data Augmentation for Multiclass Brain Tumor Detection...

引用

8th International Conference on Information technology, Information systems and Electrical engineering, ICITISEE 2024

作者： Prayogo, Rizal Dwi Nambo, Hidetaka Graduate School of Natural Science and Technology Kanazawa University Division of Electrical Engineering and Computer Science Kanazawa Japan Institute of Philosophy in Interdisciplinary Sciences Kanazawa University Faculty of Transdisciplinary Sciences Kanazawa Japan

ISBN: (纸本)9798350368970

Dealing with class imbalance is a significant issue in classification tasks that often leads to lower prediction performance. Many data augmentation methods have been suggested to tackle this problem, but their effectiveness can vary depending on the dataset. Utilizing augmentation methods can improve the learning and generalization abilities of the proposed convolutional neural network (CNN) model. However, it is challenging to find a single data augmentation method that can effectively address class imbalance problems across all scenarios. In this paper, we present a technique known as smart data augmentation (SDA) that is aimed at effectively improving classification performance by enhancing imbalanced data. SDA involves an equation that can be used to create an augmentation technique, combining different sampling methods for addressing multiclass imbalance and offering easy customization options. We perform the data balancing strategy including undersampling, oversampling, and SDA-based sampling. We evaluate our proposed framework using a CNN model to improve the detection of brain tumors in imbalanced data. Our experiments demonstrate that SDA-based sampling improves brain tumor detection performance by 3.546% in accuracy, 3.310% in precision, 3.432% in recall, and 3.371 % in Fl-score compared to the imbalanced data scenario. The findings suggest that utilizing SDA together with CNN can improve the classification performance of brain tumor detection with imbalanced data. © 2024 IEEE.

关键词： Brain tumor detection class imbalance problem convolutional neural network smart data augmentation

来源：评论

学校读者我要写书评

暂无评论

H_(2)O-balance-regulated cation-anion competitive coordination for selective elements extraction from spent lithium-ion batteries

引用

escience 2024年第6期4卷 118-126页

作者： Mingqiang Cheng Yixin Hua Qibo Zhang Qihao Li Hongda Li Ding Wang Xianshu Wang Yun Zhao Juanjian Ru Baohua Li Faculty of Metallurgical and Energy Engineering State Key Lab of Complex Nonferrous Metal Resources Clean UtilizationKunming University of Science and TechnologyKunming 650093China Institute of Materials Research Tsinghua Shenzhen International Graduate SchoolTsinghua UniversityShenzhen 518055China

Key resources necessary for lithium-ion batteries(LIBs)will deplete rapidly if recycling is not considered given the significant demand for ***,the current recovery method from spent LIBs is hindered by low efficiency,high energy consumption,and severe environmental issues,which curtail the sustainability of recycling ***,we propose a new strategy for selective elements extraction from LiCoO_(2) based on H_(2)O-balance-regulated cation-anion competitive *** lithium can be preferentially precipitated in deep eutectic solvents(DES)with an appropriate amount of H_(2)*** a preferential lithium behaviors contributes to the subsequent precise separation of transition metal elements by further adjusting the water content,thus enabling the recovery of all element and direct regeneration of DES for further spent LIB *** developed DES-based technology can achieve the ultrahigh leaching efficiency of 99.99% for Li and Co with an acceptable recovery efficiency of Li and Co(≥91.23%)and a desirable purity of recycled Li_(2)C_(2)O_(4) and CoC_(2)O_(4)·2H_(2)O(≥98.43%).These Li and Co performances are still maintained at this level even after three-time regeneration of *** methodology can also be extended to other recycling chemistries of spent LIBs and lay the theoretical foundation for the recovery of valuable metals from spent battery materials using DESs with a simple process,low energy consumption,and waste-free recycling.

关键词： battery lithium elements

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：