检索结果-内蒙古大学图书馆

6th International Conference on High Performance Computing in Asia-Pacific Region, HPC Asia 2023

作者： Kobayashi, Ryohei Fujita, Norihisa Yamaguchi, Yoshiki Boku, Taisuke Yoshikawa, Kohji Abe, Makito Umemura, Masayuki Center for Computational Sciences University of Tsukuba Japan Degree Programs in Systems and Information Engineering University of Tsukuba Japan

ISBN: (纸本)9781450398060

The complementary use of graphics processing units (GPUs) and field programmable gate arrays (FPGAs) is a major topic of interest in the high-performance computing (HPC) field. GPU-FPGA-Accelerated computing is an effective tool for multiphysics simulations, which encompass multiple physical models and simultaneous physical phenomena. Because the constituent operations in multiphysics simulations exhibit varying characteristics, accelerating these operations solely using GPUs is often challenging. Hence, FPGAs are frequently implemented for this purpose. The objective of the present study was to further improve application performance by employing both GPUs and FPGAs in a complementary manner. Recently, this approach has been applied to the radiative transfer simulation code for astrophysics known as ARGOT, with evaluation results quantitatively demonstrating the resulting improvement in performance. However, the evaluation results in question came from the use of a single node equipped with both a GPU and FPGA. In this study, we extended the GPU-FPGA-Accelerated ARGOT code to operate on multiple nodes using the message passing interface (MPI) and an FPGA-To-FPGA communication technology scheme called Communication Integrated Reconfigurable CompUting System (CIRCUS). We evaluated the performance of the ARGOT code with multiple GPUs and FPGAs under weak scaling conditions, and found it to achieve up to 12.8x speedup compared to the GPU-only execution. © 2023 Owner/Author.

关键词： Multiphysics

来源：评论

学校读者我要写书评

暂无评论

Secret key rate of continuous-variable quantum key distribution with finite codeword length

引用

Science China(information Sciences) 2023年第8期66卷 150-151页

作者： Yan FENG Runhe QIU Kun ZHANG Xue-Qin JIANG Meixiang ZHANG Peng HUANG Guihua ZENG School of Information Science and Technology Donghua University School of Electronics and Information Soochow University School of Information Engineering Yangzhou University State Key Laboratory of Advanced Optical Communication Systems and Networks Center of Quantum Sensing and Information Processing (QSIP) Shanghai Jiao Tong University

In recent years, continuous-variable quantum key distribution(CV-QKD) has been proposed as a promising alternative to the most commonly implemented discrete-variable QKD [1]. A typical CV-QKD system includes mainly tw...

关键词：

来源：评论

学校读者我要写书评

暂无评论

OUT-OF-DOMAIN UNLABELED DATA IMPROVES GENERALIZATION 12

OUT-OF-DOMAIN UNLABELED DATA IMPROVES GENERALIZATION

引用

12th International Conference on Learning Representations, ICLR 2024

作者： Saberi, Amir Hossein Movasaghinia, Mohammad Hosein Najafi, Amir Motahari, Seyed Abolfazl Heidari, Alireza Khalaj, Babak H. Department of Electrical Engineering Department of Computer Engineering Sharif Center for Information Systems and Data Science Iran Sharif Institute for Convergence Science & Technology Sharif University of Technology Tehran Iran

We propose a novel framework for incorporating unlabeled data into semi-supervised classification problems, where scenarios involving the minimization of either i) adversarially robust or ii) non-robust loss functions have been considered. Notably, we allow the unlabeled samples to deviate slightly (in total variation sense) from the in-domain distribution. The core idea behind our framework is to combine Distributionally Robust Optimization (DRO) with self-supervised training. As a result, we also leverage efficient polynomial-time algorithms for the training stage. From a theoretical standpoint, we apply our framework on the classification problem of a mixture of two Gaussians in Rd, where in addition to the m independent and labeled samples from the true distribution, a set of n (usually with n m) out of domain and unlabeled samples are given as well. Using only the labeled data, it is known that the generalization error can be bounded by ∝ (d/m)1/2. However, using our method on both isotropic and non-isotropic Gaussian mixture models, one can derive a new set of analytically explicit and non-asymptotic bounds which show substantial improvement on the generalization error compared to ERM. Our results underscore two significant insights: 1) out-of-domain samples, even when unlabeled, can be harnessed to narrow the generalization gap, provided that the true data distribution adheres to a form of the "cluster assumption", and 2) the semi-supervised learning paradigm can be regarded as a special case of our framework when there are no distributional shifts. We validate our claims through experiments conducted on a variety of synthetic and real-world datasets. © 2024 12th International Conference on Learning Representations, ICLR 2024. All rights reserved.

关键词： Polynomial approximation

来源：评论

学校读者我要写书评

暂无评论

Using Intel oneAPI for Multi-hybrid Acceleration Programming with GPU and FPGA Coupling

Using Intel oneAPI for Multi-hybrid Acceleration Programming...

引用

2024 International Conference on High Performance Computing in Asia-Pacific Region Workshops, HPC Asia 2024 Workshops

作者： Liang, Wentao Fujita, Norihisa Kobayashi, Ryohei Boku, Taisuke Degree Program in Systems and Information Engineering University of Tsukuba Ibaraki Tsukuba Japan Center for Computational Sciences University of Tsukuba Ibaraki Tsukuba Japan

ISBN: (纸本)9798400716522

Intel oneAPI is a programming framework that accepts various accelerators such as GPUs, FPGAs, and multi-core CPUs, with a focus on HPC applications. Users can apply their code written in a single language, DPC++, to this heterogeneous programming environment. However, in practice, it is not easy to apply to different accelerators, especially for non-Intel devices such as NVIDIA and AMD GPUs. We have successfully constructed a oneAPI environment set to utilize the single DPC++ programming to handle true multi-hetero acceleration including NVIDIA GPU and Intel FPGA simultaneously. In this paper, we will show how this is done and what kind of applications can be targeted. © 2024 Copyright held by the owner/author(s).

关键词： Multiphysics

来源：评论

学校读者我要写书评

暂无评论

Auditory VR Generative System for Non-Experts to Reproduce Human Memories Through Natural Language Interactions 23

Auditory VR Generative System for Non-Experts to Reproduce H...

引用

2023 SIGGRAPH Asia Posters, SA 2023

作者： Yamauchi, Yuta Ino, Keiko Zempo, Keiichi College of Engineering Systems University of Tsukuba Tsukuba Japan National Institute of Mental Health National Center of Neurology and Psychiatry Kodaira Japan Institute of Systems and Information Engineering University of Tsukuba Tsukuba Japan

ISBN: (纸本)9798400703133

We propose an automatic auditory VR generative system based on natural language input and attempt to apply it to VR exposure therapy, a promising treatment for Post-Traumatic Stress Disorder (PTSD). The system consisted of the user interface, developed based on the Large Language Model (LLM), the auditory event dataset, which has metadata of "subject" and "verb" and spatial audio generator. © 2023 Copyright held by the owner/author(s).

关键词： User interfaces

来源：评论

学校读者我要写书评

暂无评论

Deep learning application in diagnosing breast cancer recurrence

引用

Multimedia Tools and Applications 2025年第13期84卷 12265-12297页

作者： Jam, Zeinab Albadvi, Amir Atashi, Alireza Department of Information Technology Engineering-Industrial and Systems Engineering Faculty Tarbiat Modares University Tehran Iran Medical Informatics Department Breast Cancer Research Center Motamed Cancer Institute ACECR Tehran Iran Department of Digital Health School of Medicine Tehran University of Medical Sciences Tehran Iran

Patients' lives can always be saved when diseases, especially special diseases, are detected early. The chances of a patient surviving can be increased by early detection. Breast cancer is one of the deadliest and common cancers. After recovering from breast cancer, patients are always worried about recurrence and return. The use of modern technology, however, can help predict disease recurrence at an early stage, allowing patients to receive treatment sooner. Significant strides have been achieved in deep learning, demonstrating strong performance in handling unstructured data challenges. However, when it comes to predicting tabular data, deep learning hasn't quite matched its success with unstructured data. Presently, ensemble models relying on gradient-boosted decision trees (GBDT) are frequently favored for tabular data prediction tasks. Typically, these GBDT-based models outshine deep learning approaches. Many novel deep learning techniques are emerging for handling tabular data. TabNet, for instance, mirrors decision tree feature selection within a neural network framework. AutoInt addresses high dimensionality by condensing data through embedding layers. Tab Transformer adapts the transformer model, generating text representations for categorical attributes. Despite their innovation, these methods remain less recognized compared to those for image and text data processing. In this study, 158 different characteristics of 5142 breast cancer patients from 1997 to 2019 were examined. We aim to evaluate deep learning techniques effectiveness in detecting breast cancer recurrence. Through examination of evaluation metrics, it becomes evident that deep learning approaches applied to tabular data surpass traditional machine learning algorithms, even when dealing with imbalanced datasets. Ultimately, the results derived from each algorithm analyzed and concluded with a review and comparison of the findings. © The Author(s), under exclusive licence to Springer Science+

关键词： Diseases

来源：评论

学校读者我要写书评

暂无评论

Supervised Local Training With Backward Links for Deep Neural Networks

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第2期5卷 854-867页

作者： Guo, Wenzhe Fouda, Mohammed E. Eltawil, Ahmed M. Salama, Khaled Nabil King Abdullah University of Science and Technology Division of Computer Electrical and Mathematical Sciences and Engineering Thuwal23955 Saudi Arabia University of California Center for Embedded and Cyber-Physical Systems IrvineCA92612 United States

The restricted training pattern in the standard backpropagation (BP) requires end-to-end error propagation, causing large memory costs and prohibiting model parallelization. Existing local training methods aim to resolve the training obstacles by completely cutting off the backward path between modules and isolating their gradients. These methods prevent information exchange between modules and result in inferior performance. This work proposes a novel local training algorithm, BackLink, which introduces intermodule backward dependence and facilitates information to flow backward along with the network. To preserve the computational advantage of local training, BackLink restricts the error propagation length within the module. Extensive experiments performed in various deep convolutional neural networks demonstrate that our method consistently improves the classification performance of local training algorithms over other methods. For example, our method can surpass the conventional greedy local training method by 6.45% in accuracy in ResNet32 classifying CIFAR100 and recent work by 2.58% in ResNet110 classifying STL-10 with much lower complexity. Analysis of computational costs reveals that small overheads are incurred in GPU memory costs and runtime on multiple GPUs. Our method can lead up to a 79% reduction in memory cost and 52% in simulation runtime in ResNet110 compared to the standard BP. Therefore, our method could create new opportunities for improving training algorithms toward better efficiency for real-time learning applications. © 2023 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Biosignal-based Control of a Robotic Gait Training Lifter 46

Biosignal-based Control of a Robotic Gait Training Lifter

引用

46th Annual International Conference of the IEEE engineering in Medicine and Biology Society, EMBC 2024

作者： Zhang, Xiaoxi Kadone, Hideki Konishi, Tomohiro Chen, Yang Hassan, Modar Suzuki, Kenji University of Tsukuba Artificial Intelligence Laboratory Tsukuba Japan University of Tsukuba Hospital Center for Innovative Medicine and Engineering Japan University of Tsukuba Institute of Systems and Information Engineering Tsukuba Japan

ISBN: (纸本)9798350371499

In this paper, we present a robotic walker that aims to encourage the patient's voluntary movement by enabling intention-based control of the mobile base. We proposed two variants of biosignal-based control methods for the robotic gait training lifter (GTL): one method to control the speed of the walker directly from Electromyography (EMG) signals of the muscles of the lower limb, and the other method to control the acceleration of the walker from EMG. We implemented these methods in a robotic walker and conducted experiments with one healthy male and one female participant to evaluate the feasibility of the proposed method. The results showed that the velocity of the human and walker can be synchronized with both controllers in the walking phase. The proposed method allows users to use together with robotic exoskeletons for rehabilitation sessions and has the potential to promote the active participation of the patient. © 2024 IEEE.

关键词： Exoskeleton (Robotics)

来源：评论

学校读者我要写书评

暂无评论

Innovation Competitions in STEM Education: A Comprehensive Analysis of Attributes and Student Experiences 14

Innovation Competitions in STEM Education: A Comprehensive A...

引用

14th IEEE Integrated STEM Education Conference, ISEC 2024

作者： Prince, Alexa Kulturel-Konak, Sadan Konak, Abdullah Pennsylvania State University University Park United States Management Information Systems Division of Engineering Business and Technology Pennsylvania State University Berks United States Information Sciences and Technology Division of Engineering Business and Technology Pennsylvania State University Berks United States

ISBN: (纸本)9798350352801

In order to gain technical and professional experience, many students in STEM fields participate in extracurricular activities, called such as Innovation Competition and Programs (ICPs). ICPs can offer students a hands-on experience in their field, fostering creativity and entrepreneurship while also solving problems as a team. While the literature emphasizes the many benefits that ICPs offer and examples of different types, there is not yet a broad discussion of different ICP attributes, or many documented student testimonials. This paper aims to establish a precise categorization of the various attributes found in such programs and examine actual student ICP experiences to understand how to adjust these attributes in order to provide students with the most positive experience possible. It is helpful to outline the potential attributes of ICPs due to their open-ended and university-dependent nature. In addition, student interview data provides insight into the overall experiences that students have had when participating in these programs, allowing connections to be made between student goals and the structure of the programs themselves. © 2024 IEEE.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Enhanced In-Flight Connectivity for Urban Air Mobility via LEO Satellite Networks 30

Enhanced In-Flight Connectivity for Urban Air Mobility via L...

引用

30th IEEE International Conference on Telecommunications, ICT 2024

作者： Biswas, Karnika Ghazzai, Hakim Khanfor, Abdullah Sboui, Lokman CEMSE Division Thuwal Saudi Arabia College of Computer Science and Information Systems Najran University Computer Science Department Najran Saudi Arabia University of Québec Systems Engineering Department Montréal Canada

ISBN: (纸本)9798350356694

Urban Air Mobility (UAM) is the envisioned future of inter-city aerial transportation. This paper presents a novel, in-flight connectivity link allocation method for UAM, which dynamically switches between terrestrial cellular and Low Earth Orbit (LEO) satellite networks based on real-time conditions. Our approach prefers cellular networks for cost efficiency, switching to LEO satellites under poor cellular conditions to ensure continuous UAM connectivity. By integrating real-time metrics like signal strength, network congestion, and flight trajectory into the selection process, our algorithm effectively balances cost, minimum data rate requirements, and continuity of communication. Numerical results validate minimization of data-loss while ensuring an optimal selection from the set of available above-threshold data rates at every time sample. Furthermore, insights derived from our study emphasize the importance of hybrid connectivity solutions in ensuring seamless, uninterrupted communication for future urban aerial vehicles. © 2024 IEEE.

关键词： Air mobility

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：