检索结果-内蒙古大学图书馆

Performance Improvement of a Virtual Network Embedding Algorithm based on Temporal-difference Learning by Resource-Constraint-Aware Candidate Solution Selection

引用

IEIE Transactions on Smart Processing and Computing 2024年第2期13卷 158-166页

作者： Fukushima, Yukinobu Sagawa, Yuta Tarutani, Yuya Yokohira, Tokumi Faculty of Natural Science and Technology Okayama University Okayama Okayama city Japan Graduate school of Natural Science and Technology Okayama University Okayama Okayama City Japan Faculty of Interdisciplinary Science and Engineering in Health Systems Okayama University Okayama Okayama city Japan

In this paper, we tackle a virtual network embedding problem in network virtualization. For this problem, an algorithm (VNE-TD) based on temporal difference learning has been proposed. VNE-TD, however, does not consider the node and link resource constraints in selecting the candidate solutions. Therefore, when attempting to construct a virtual network, the embedding of a virtual network may fail due to insufficient resources. In this paper, we modify VNE-TD to select only those candidate solutions that satisfy the node and link resource constraints. We add a function to check the satisfiability of the node and link resource constraints to VNE-TD. The simulation results show that our modified methods reduce the blocking ratio of virtual network requests by up to 80% compared to VNE-TD. © 2024 Institute of Electronics Engineers of Korea. All rights reserved.

关键词： Network embeddings

来源：评论

学校读者我要写书评

暂无评论

Food Volume Prediction and Classification using Multi-Task Learning 6

Food Volume Prediction and Classification using Multi-Task L...

引用

6th World Symposium on Artificial Intelligence, WSAI 2024

作者： Sari, Yuita Arum Nakazawa, Atsushi Gofuku, Akio Graduate School of Interdisciplinary Science and Engineering in Health Systems Okayama University Okayama Japan Brawijaya University Faculty of Computer Science Malang Indonesia

ISBN: (纸本)9798350374162

Food volume prediction is critical for calculating calories in food analysis and nutritional evaluation. However, obtaining data and measuring volume from the 3D sensor camera concurrently is time-consuming. Additionally, high computing power is necessary to compute the volume. Therefore, machine learning techniques are employed to extract specific data collected from the data collection process for learning purposes to save time. In this study, we employ Multi-Task Learning to predict volume and categorize food due to their interdependence simultaneously. We evaluate several backbones and find that ResNet50 attains the best performance with an MAE of 4.87 for volume prediction and 100% food classification accuracy. Our proposed method showcases outstanding results, aided by an efficient volume measurement methodology © 2024 IEEE.

关键词： food volume prediction machine learning multi-task learning

来源：评论

学校读者我要写书评

暂无评论

Improved waterproof performance of ultra-thin Pt film hydrogen sensors with a glass package 10

Improved waterproof performance of ultra-thin Pt film hydrog...

引用

10th IEEE Asia-Pacific Conference on Applied Electromagnetics, APACE 2024

作者： Oh, Yuki Nakayama, Haruhito Wang, Jin Kiwa, Toshihiko Okayama University Graduate School of Interdisciplinary Science and Engineering in Health Systems Okayama Japan

ISBN: (纸本)9798350364163

With the increasing importance of hydrogen energy in recent years, highly sensitive hydrogen sensors are required to detect hydrogen gas leaks at an early stage. Our group has developed the ultra-thin platinum (Pt) film hydrogen sensors that can operate at room temperature. In this study, we have developed a waterproof package for the ultra-thin Pt film hydrogen sensors to ensure they are less affected by ambient conditions such as rain. As a result, the waterproof performance of the package meets the protection class 7 of the IP Code. © 2024 IEEE.

关键词： Waterproofing

来源：评论

学校读者我要写书评

暂无评论

Measuring Biotin Concentrations for Quantitative Magnetic Immunoassay Using a Femtosecond Laser 10

Measuring Biotin Concentrations for Quantitative Magnetic Im...

引用

10th IEEE Asia-Pacific Conference on Applied Electromagnetics, APACE 2024

作者： Asayama, Shu Kako, Sakuya Yamashita, Kei Wang, Jin Kiwa, Toshihiko Okayama University Graduate School of Interdisciplinary Science and Engineering in Health Systems Okayama Japan

ISBN: (纸本)9798350364163

Magnetic immunoassay (MIA), a medical application that utilizes magnetic nanoparticles (MNPs), is characterized by its ability to provide quantitative evaluation of antigen levels in a short time with high accuracy, without requiring a washing process. However, a challenge with this technique is that the magnetic signal can decrease due to the aggregation of the MNPs. In this study, we attempted to redisperse the aggregates by irradiating the sample with a laser. The results showed that laser irradiation produced a significantly greater change in signal change compared to the case without laser irradiation. Therefore, it was suggested that laser irradiation can suppress the magnetic signal decrease due to aggregation and enable accurate estimation of the antigen content in the sample. © 2024 IEEE.

关键词： Femtosecond lasers

来源：评论

学校读者我要写书评

暂无评论

Precise Cancellation of Magnetic Signal Using a Movable Ferromagnetic Rod for Magnetic Nanoparticles Measurements 10

Precise Cancellation of Magnetic Signal Using a Movable Ferr...

引用

10th IEEE Asia-Pacific Conference on Applied Electromagnetics, APACE 2024

作者： Ogawa, Rikako Yamashita, Kei Wang, Jin Kiwa, Toshihiko Okayama University Graduate School of Interdisciplinary Science and Engineering in Health Systems Okayama Japan

ISBN: (纸本)9798350364163

In this study, a new magnetic signal cancellation method using a movable ferromagnetic rod is proposed to reduce magnetic signal noise in the measurement of magnetic nanoparticles (MNPs). Noise from the ambient magnetic field and from the instrument itself has a significant effect on the measurement results of MNPs. Therefore, this study developed a technique to cancel this noise using a movable rod made of ferromagnetic material. Experimental results confirm that proposed technique reduces noise by 10.2 mV and improves the accuracy of MNPs measurements. This technique is expected to enable more precise measurement and diagnosis in various medical applications using MNPs in the future. © 2024 IEEE.

关键词： Magnetic nanoparticles

来源：评论

学校读者我要写书评

暂无评论

Development of an AC Magnetic Characterization System and Its Application to Magnetic Immunoassay 10

Development of an AC Magnetic Characterization System and It...

引用

10th IEEE Asia-Pacific Conference on Applied Electromagnetics, APACE 2024

作者： Korenaga, Rikuya Yamashita, Kei Naito, Kosuke Wang, Jin Kiwa, Toshihiko Okayama University Graduate School of Interdisciplinary Science and Engineering in Health Systems Okayama Japan

ISBN: (纸本)9798350364163

Magnetic nanoparticles (MNPs) are nanometer-sized magnetic particles that are expected to have various medical applications. To achieve highly sensitive testing, it is necessary to evaluate the AC magnetic properties of MNPs under various conditions. Commercially available systems use SQUID sensors and have excellent detection sensitivity. However, they have problems such as the need for expensive refrigerants and the large equipment to maintain the superconducting state. Therefore, in this research, we aimed to develop a low-cost and compact AC magnetic characterization system consisting of two lock-in amplifier circuits. Additionally, we conducted an experiment to detect biotin-avidin binding reaction, which is the target of conventional optical methods, and investigated the effect of immunoreaction on the AC magnetic properties of MNPs. As a result of the experiment, we were able to measure the AC magnetic properties of MNPs in the liquid phase using the developed system. Furthermore, in immunoreaction experiments, it was confirmed that the signal intensity of MNPs decreased as the biotin concentration increased. This is thought to be due to the antigen-antibody reaction between biotin and the streptavidin antibody modified on the MNPs surface, which caused the MNPs to aggregate and change the AC magnetic properties. These results suggest that the developed system is applicable to quantitative liquid-phase immunoassays. © 2024 IEEE.

关键词： Antigen-antibody reactions

来源：评论

学校读者我要写书评

暂无评论

Development of Viscosity Measurement System Utilizing the Frequency Characteristics of Magnetic Nanoparticles 10

Development of Viscosity Measurement System Utilizing the Fr...

引用

10th IEEE Asia-Pacific Conference on Applied Electromagnetics, APACE 2024

作者： Yamashita, Kei Naito, Kosuke Korenaga, Rikuya Wang, Jin Kiwa, Toshihiko Okayama University Graduate School of Interdisciplinary Science and Engineering in Health Systems Okayama Japan

ISBN: (纸本)9798350364163

Viscosity measurement is a crucial parameter for understanding the properties of substances and has applications in medical fields such as coagulation measurement. In this study, we aimed to measure viscosity by utilizing the dependence of the frequency characteristics of the AC magnetic susceptibility of magnetic nanoparticles on viscosity. First, we developed a system to measure the AC magnetic susceptibility of magnetic nanoparticles using the gain-phase method. Then, assuming a log-normal distribution of the particle size of the magnetic nanoparticles, we analyzed the viscosity from the imaginary part of the frequency characteristics of the AC magnetic susceptibility obtained by the developed system. The correlation between the analyzed viscosity and the viscosity obtained from a viscometer and a densitometer was y = 0.607x + 0.471 (R2 = 0.993), indicating a strong linear correlation. This result suggested that the developed system can measure the solvent viscosity through calibration. © 2024 IEEE.

关键词： Viscosity measurement

来源：评论

学校读者我要写书评

暂无评论

Experiment Design for EEG-based Continuous Action Sense of Agency 15

Experiment Design for EEG-based Continuous Action Sense of A...

引用

15th Biomedical engineering International Conference, BMEiCON 2023

作者： Widodo, Arif Nakazawa, Atsushi Gofuku, Akio Okayama University Graduate School of Interdisciplinary Science and Engineering in Health Systems Okayama Japan

ISBN: (纸本)9798350345247

Previous research on EEG-based Sense of Agency (SoA) has identified frequency bands and locations associated with SoA. However, the time appearance of SoA on EEG is rarely discussed. Other studies have linked the relationship between SoA and error-related negativity (ERN), which gives hope for studying SoA in the time domain. As a result, EEG data with time markers that can be analyzed in the time domain is necessary. This research aims to create an experiment game that can provide SoA stimulus while also recording all of the subject's response activities. The initial experiment discovered that calculating the subject's movements can estimate the feeling of being in control. Furthermore, as a result of losing agency, the subject's response time is longer in bad moves than in good moves. © 2023 IEEE.

关键词： continuous action EEG experiment design sense of agency

来源：评论

学校读者我要写书评

暂无评论

Complex Emotion Estimation using Analysis-by-Synthesis of Facial Expression Images

引用

IEEE Access 2025年 13卷 88731-88746页

作者： Khine, Win Shwe Sin Siritanawan, Prarinya Kotani, Kazunori Japan Advanced Institute of Science and Technology School of Information Science Japan Shinshu University Informatics and Interdisciplinary Systems Division Graduate School of Science and Technology Department of Engineering Nagano Japan Kanazawa University Faculty of Transdisciplinary Science Institute of Philosophy in Interdisciplinary Science Ishikawa Japan

This research proposes an approach for recognizing facial expressions for complex emotion estimation through the utilization of generated facial images. The proposed approach consists of two main parts: facial image generation and facial expression recognition. In the first part, we introduce Conditioned Emotion Generative Adversarial Networks (cEmoGANs) to synthesize images that convey complex facial expressions. Unlike previous methods, our generative model maintains face identity information and expresses various types of complex emotions. This capability encourages the generator to have control over the image generation process, resulting in enhanced image quality, reduced distortion, and increased diversity of generated images. The second part involves the design of a multiple-label classification based on a convolutional neural network trained on the complex facial expression images generated from the first part, which is employed in the recognition of complex facial expressions for emotion estimation. Our model, using images generated by cEmoGANs, demonstrates a notable performance surpassing the capabilities of the previous models in comparative evaluations. © 2013 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Image Integrity Analysis under Defense Approaches Against Adversarial Patches

Image Integrity Analysis under Defense Approaches Against Ad...

引用

2025 International Workshop on Advanced Imaging Technology, IWAIT 2025

作者： Siritanawan, Prarinya Chunhachatchawhankhun, Kasidit Kotani, Kazunori Graduate School of Science and Technology Department of Engineering Informatics and Interdisciplinary Systems Division Shinshu University Nagano Japan Division of Transdisciplinary Science JAIST Ishikawa Japan Faculty of Transdisciplinary Science Institute of Philosophy in Interdisciplinary Science Kanazawa University Ishikawa Japan

ISBN: (数字)9781510688131

ISBN: (纸本)9781510688124

Adversarial patches pose a significant threat to image recognition systems, exploiting small, maliciously crafted patterns to mislead models into incorrect *** traditional adversarial examples, adversarial patches operate under real-world constraints, making defense against them a challenging *** study introduces an alternative defense strategy that employs partial masking to mitigate the impact of adversarial patches without relying on precise patch *** global masking patterns were evaluated to determine the best balance between adversarial robustness and image *** proposed method effectively reduces attack success rates while maintaining reasonable classification accuracy, demonstrating its potential to ensure protection in real-world applications without dependency on the detection methods. © 2024 SPIE.

关键词： Defense and security Image classification Image restoration Image analysis Visualization Image processing Data modeling systems modeling

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：