检索结果-内蒙古大学图书馆

International Conference on Pattern Recognition

作者： Michele Cancilla Laura Canalini Federico Bolelli Stefano Allegretti Salvador Carrión Roberto Paredes Jon A. Gómez Simone Leo Marco Enrico Piras Luca Pireddu Asaf Badouh Santiago Marco-Sola Lluc Alvarez Miquel Moreto Costantino Grana Università degli Studi di Modena e Reggio Emilia Italy PRHLT Research Center Universitat Politècnica de València Spain Data-intensive Computing Group CRS4 Italy Barcelona Supercomputing Center Spain Universitat Autònoma de Barcelona Spain Universitat Politècnica de Catalunya Spain

Given the overwhelming impact of machine learning on the last decade, several libraries and frameworks have been developed in recent years to simplify the design and training of neural networks, providing array-based programming, automatic differentiation and user-friendly access to hardware accelerators. None of those tools, however, was designed with native and transparent support for Cloud computing or heterogeneous High-Performance computing (HPC). The DeepHealth Toolkit is an open source Deep Learning toolkit aimed at boosting productivity of data scientists operating in the medical field by providing a unified framework for the distributed training of neural networks, which is able to leverage hybrid HPC and cloud environments in a transparent way for the user. The toolkit is composed of a Computer Vision library, a Deep Learning library, and a front-end for non-expert users; all of the components are focused on the medical domain, but they are general purpose and can be applied to any other field. In this paper, the principles driving the design of the DeepHealth libraries are described, along with details about the implementation and the interaction between the different elements composing the toolkit. Finally, experiments on common benchmarks prove the efficiency of each separate component and of the DeepHealth Toolkit overall.

关键词： Deep learning Training Computer vision Cloud computing Pipelines Neural networks Writing

来源：评论

学校读者我要写书评

暂无评论

Explainable AI: Definition and attributes of a good explanation for health AI

arXiv

引用

arXiv 2024年

作者： Kyrimi, Evangelia McLachlan, Scott Wohlgemut, Jared M. Perkins, Zane B. Lagnado, David A. Marsh, William Gimson, Alexander Shafti, Ali Ercole, Ari Banerjee, Amitava Glocker, Ben Schafer, Burkhard Gatsonis, Constantine Grosan, Crina Sent, Danielle Berman, David S. Glass, David O'Regan, Declan P. Letsios, Dimitrios Morrissey, Dylan Pisirir, Erhan Leofante, Francesco Soyel, Hamit Williamson, Jon Grieman, Keri Dube, Kudakwashe Mardsen, Max Nagendran, Myura Tai, Nigel Kostopoulou, Olga Jones, Owain Curzon, Paul Stoner, Rebecca S. Tandle, Sankalp Joshi, Shalmali Mossadegh, Somayyeh Buijsman, Stefan Miller, Tim Madai, Vince Istvan School of Electronic Engineering and Computer Science Queen Mary University of London London United Kingdom School of Nursing Midwifery and Palliative Care King's College London London United Kingdom Centre for Trauma Sciences Blizard Institute Queen Mary University of London London United Kingdom Royal London Hospital Barts Health NHS Trust London United Kingdom Department of Experiment Psychology University College London London United Kingdom Cambridge Liver Unit Addenbrooke's Hospital Cambridge University Hospitals Hills Road Box 210 Cambridge United Kingdom University of Cambridge United Kingdom Cambridge Consultants Ltd. United Kingdom Division of Anaesthesia Cambridge Centre for AI in Medicine University of Cambridge Cambridge United Kingdom Magdelene College University of Cambridge United Kingdom Institute of Health Informatics University College London London United Kingdom Department of Computing Imperial College London London United Kingdom SCRIPT Centre for IT and IP Law School of Law University of Edinburgh United Kingdom Center for Statistical Sciences School of Public Health Brown University Providence United States Research Group Artificial Intelligence HU University of Applied Sciences Utrecht Netherlands Jheronimus Academy of Data Science Tilburg University Eindhoven University of Technology s-Hertogenbosch Netherlands Centre for Theoretical Physics Queen Mary University of London London United Kingdom School of Computing Ulster University Belfast United Kingdom MRC Laboratory of Medical Sciences Imperial College London London United Kingdom Department of Informatics King's College London United Kingdom Physiotherapy Department Barts Health NHS Trust London United Kingdom Sport and Exercise Medicine Queen Mary University of London London United Kingdom School of Electronic Engineering and Computer Science Queen Mary University of London United Kingdom Department of Computing Imperial College London United Kingdom Departmen

Proposals of artificial intelligence (AI) solutions based on increasingly complex and accurate predictive models are becoming ubiquitous across many disciplines. As the complexity of these models grows, transparency and users' understanding often diminish. This suggests that accurate prediction alone is insufficient for making an AI-based solution truly useful. In the development of healthcare systems, this introduces new issues related to accountability and safety. Understanding how and why an AI system makes a recommendation may require complex explanations of its inner workings and reasoning processes. Although research on explainable AI (XAI) has significantly increased in recent years and there is high demand for XAI in medicine, defining what constitutes a good explanation remains ad hoc, and providing adequate explanations continues to be challenging. To fully realize the potential of AI, it is critical to address two fundamental questions about explanations for safety-critical AI applications, such as health-AI: (1) What is an explanation in health-AI? and (2) What are the attributes of a good explanation in health-AI? In this study, we examined published literature and gathered expert opinions through a two-round Delphi study. The research outputs include (1) a definition of what constitutes an explanation in health-AI and (2) a comprehensive list of attributes that characterize a good explanation in health-AI. © 2024, CC BY.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Enhanced Usability of Managing Workflows in an Industrial data Gateway

Enhanced Usability of Managing Workflows in an Industrial Da...

引用

IEEE International Conference on e-Science and Grid computing

作者： Gary A. McGilvary Malcolm Atkinson Sandra Gesing Alvaro Aguilera Richard Grunzke Eva Sciacca Edinburgh Data-Intensive Research Group The University of Edinburgh Center for Research Computing University of Notre Dame Indiana United States Center for Information Services and High Performance Computing (ZIH) Technische Universitat Dresden Germany INAF-Osservatorio Astrofisico di Catania italy

The Grid and Cloud User Support Environment (gUSE) enables users convenient and easy access to grid and cloud infrastructures by providing a general purpose, workflow-oriented graphical user interface to create and run workflows on various Distributed computing Infrastructures (DCIs). Its arrangements for creating and modifying existing workflows are, however, non-intuitive and cumbersome due to the technologies and architecture employed by gUSE. In this paper, we outline the first integrated web-based workflow editor for gUSE with the aim of improving the user experience for those with industrial data workflows and the wider gUSE community. We report initial assessments of the editor's utility based on users' feedback. We argue that combining access to diverse scalable resources with improved workflow creation tools is important for all big data applications and research infrastructures.

关键词： Java Concrete Usability databases Context Security Libraries

来源：评论

学校读者我要写书评

暂无评论

Latest advances in distributed, parallel, and graphic processing unit accelerated approaches to computational biology

Latest advances in distributed, parallel, and graphic proces...

引用

作者： Merelli, Ivan Pérez-Sánchez, Horacio Gesing, Sandra D'Agostino, Daniele Bioinformatics Research Unit Institute for Biomedical Technologies National Research Council of Italy Segrate Milan Italy Murcia Spain Data-Intensive Research Group University of Edinburgh Edinburgh United Kingdom Advanced Computing Systems and High Performance Computing Group Institute for Applied Mathematics and Information Technologies National Research Council of Italy Genoa Italy

The special issue of Concurrency And Computation: Practice And Experience deals with latest advances in distributed, parallel, and graphic processing unit accelerated approaches to computational biology. This trend is motivated by the lightening improvement of novel molecular biology high-throughput technologies, such as next generation sequencing, which allow the analysis of inter personal variations in genomics and transcriptomics. This also results in the development of mass spectrometry techniques for proteomics and metabolomics profiles. Many projects have been undertaken in a number of countries and universities to achieve these objectives. The D-Grid project MoSGrid is one such project that offers a complete solution for the molecular simulation community supporting HPC infrastructures via a web-based science gateway.

关键词： Graphics processing unit

来源：评论

学校读者我要写书评

暂无评论

Distributing Power Grid State Estimation on HPC Clusters - A System Architecture Prototype

Distributing Power Grid State Estimation on HPC Clusters - A...

引用

IEEE International Symposium on Parallel and Distributed Processing Workshops and Phd Forum (IPDPSW)

作者： Yan Liu Wei Jiang Shuangshuang Jin Mark Rice Yousu Chen Data Intensive Computing Group Pacific Northwest National Laboratory Richland WA USA Department of Computer Science and Engineering Ohio State University Columbus OH USA Electrical Power Systems Engergy Pacific Northwest National Laboratory Richland WA USA

ISBN: (纸本)9781467309745

The future power grid is expected to further expand with highly distributed energy sources and smart loads. The increased size and complexity lead to increased burden on existing computational resources in energy control centers. Thus the need to perform real-time assessment on such systems entails efficient means to distribute centralized functions such as state estimation in the power system. In this paper, we present our experience of prototyping a system architecture that connects distributed state estimators individually running parallel programs to solve non-linear estimation procedure. Through our experience, we highlight the needs of integrating the distributed state estimation algorithm with efficient partition and data communication tools so that distributed state estimation has low overhead compared to the centralized solution. We build a test case based on the IEEE 118 bus system and partition the state estimation of the whole system model to available HPC clusters. The measurement from the test bed demonstrates the low overhead of our solution.

关键词： State estimation Computer architecture Size measurement Distributed databases Power grids Partitioning algorithms

来源：评论

学校读者我要写书评

暂无评论

It takes glue to tango: MeDICi integration framework's data-intensive computing pipeline

引用

Scientific computing 2008年第7期25卷 16-19页

作者： Gorton, Ian Oehmen, Christopher S. McDermott, Hason E. PNNL's Data Intensive Computing Program United States Laboratory's Bioinformatics and Computational Biology Group Fundamental and Computational Sciences Directorate United States

Researchers at the Department of Energy's (DOE) Pacific Northwest National Laboratory (PNNL) in Richland, WA, are creating computing environments for biologists that seamlessly integrate collections of data and computational resources. MeDICi is an evolving middleware platform for building complex, high-performance analytical applications. MIF components are constructed using Java programming interfaces that support inter-component communication using asynchronous messaging. Local components execute inside the MIF container. Remote components create distributed solutions and integrate with non-Java code. Mule provides the MIF container environment. MIF extends the Mule interface to make component and pipeline construction easier and to create an encapsulation device for component creation. The MIF interface is agnostic of the underlying Java messaging platform. This allows deployments to configure MIF applications using technologies that meet individual quality-of-service requirements.

关键词： Containers

来源：评论

学校读者我要写书评

暂无评论

Radiation-coupled front-tracking simulations for laser-driven shock experiments

Radiation-coupled front-tracking simulations for laser-drive...

引用

作者： Zhang, Yongmin Drake, R. Paul Glimm, James Grove, John W. Sharp, David H. Department of Applied Mathematics and Statistics University at Stony Brook Stony Brook NY 11974-3600 United States University of Michigan 2455 Hayward St. Ann Arbor MI 48105 United States Center for Data Intensive Computing Brookhaven National Laboratory Upton NY 11793-6000 United States Continuum Dynamics Group Computer and Computational Science Division Los Alamos National Laboratory Los Alamos NM 87545 United States Complex Systems Group Theoretical Division Los Alamos National Laboratory Los Alamos NM 87545 United States

The purpose of this paper is to develop a numerical algorithm to track the preheat interface motion driven by radiation transfer in high-intensity laser experiments. Our front-tracking algorithm is coupled to a radiation process through an inter-package coupling by connecting the output from a radhydro code HYADES to the input of FronTier code of front tracking. Our coupled algorithm is validated by comparing simulation results from both codes in both low and high radiation cases. Significant interface motion and deformation of the harmonic perturbation due to radiation preheat are observed in high radiation heat case. © 2005 Elsevier Ltd. All rights reserved.

关键词： 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：