15 Hits in 2.3 sec

A Data Ecosystem to Support Machine Learning in Materials Science [article]

Ben Blaiszik, Logan Ward, Marcus Schwarting, Jonathon Gaff, Ryan Chard, Daniel Pike, Kyle Chard, Ian Foster
2019 arXiv   pre-print
of new data across the ecosystem, and the connecting of data with materials-specific machine learning models.  ...  We use examples to show how MDF and DLHub capabilities can be leveraged to link data with machine learning models and how users can access those capabilities through web and programmatic interfaces.  ...  The frst combines data hosted and indexed within MDF Publish and Discover with ML models published to DLHub to rapidly predict band gap based on an input image.  ... 
arXiv:1904.10423v1 fatcat:gibssrxzxbaqxdjsfa4ld4gd2a

DLHub: Model and Data Serving for Science [article]

Ryan Chard, Zhuozhao Li, Kyle Chard, Logan Ward, Yadu Babuji, Anna Woodard, Steve Tuecke, Ben Blaiszik, Michael J. Franklin, Ian Foster
2018 arXiv   pre-print
Here we present the Data and Learning Hub for science (DLHub), a multi-tenant system that provides both model repository and serving capabilities with a focus on science applications.  ...  Unlike other model serving frameworks, DLHub can store and serve any Python 3-compatible model or processing function, plus multiple-function pipelines.  ...  ACKNOWLEDGMENTS This work was supported in part by Laboratory Directed Research and Development (LDRD) funding from Argonne National Laboratory and the RAMSES project, both from the U.S.  ... 
arXiv:1811.11213v1 fatcat:zmmandowgnco7n3b5rqd3nao5q

Confluence of Artificial Intelligence and High Performance Computing for Accelerated, Scalable and Reproducible Gravitational Wave Detection [article]

E. A. Huerta, Asad Khan, Xiaobo Huang, Minyang Tian, Maksim Levental, Ryan Chard, Wei Wei, Maeve Heflin, Daniel S. Katz, Volodymyr Kindratenko, Dawei Mu, Ben Blaiszik (+1 others)
2020 arXiv   pre-print
We develop a workflow that connects the Data and Learning Hub for Science (DLHub), a repository for publishing machine learning models, with the Hardware Accelerated Learning (HAL) deep learning computing  ...  In this work, we demonstrate how connecting recently deployed DOE and NSF-sponsored cyberinfrastructure allows for new ways to publish models, and to subsequently deploy these models into applications  ...  DLHub is a system that provides model repository and model serving facilities, in particular for machine-learning models with science applications.  ... 
arXiv:2012.08545v1 fatcat:da43bh6mxrbuhkxymyvqn6danm

Federated Function as a Service for eScience

Yadu Babuji, Josh Bryan, Ryan Chard, Kyle Chard, Ian Foster, Ben Galewsky, Daniel S. Katz, Zhuozhao Li
2021 Zenodo  
This deposit contains a preprint of a poster paper, and the poster itself, for "Federated Function as a Service for eScience" as submitted and peer-reviewed by the 17th IEEE International Conference on  ...  This is in part due to barriers such as the need to access large research data, diverse hardware requirements, monolithic code bases, and existing systems available to researchers.  ...  The Data and Learning Hub for science for science (DLHub) [6] is a multi-tenant system that provides both a model repository and model serving capabilities with a focus on science applications.  ... 
doi:10.5281/zenodo.5519583 fatcat:7avwlncw4ngfje4ar4pdaabzzy

Infrastructure for Analysis of Large Microscopy and Microanalysis Data Sets

Jingrui Wei, Carter Francis, Dane Morgan, KJ Schmidt, Aristana Scourtas, Ian Foster, Ben Blaiszik, Paul M. Voyles
2022 Microscopy and Microanalysis  
Analysis methods span a range from physically-motivated approaches with no adjustable parameters, through unsupervised machine learning (ML) methods which require careful hyperparameter optimization, to  ...  Foundry unifies two existing services: the Materials Data Facility, which offers large scale storage and permanent accessibility to data, and DLHub, which publishes and executes ML models.  ... 
doi:10.1017/s1431927622011539 fatcat:3fjc7fxadfbopmvisnj53i4pby

MLModelCI: An Automatic Cloud Platform for Efficient MLaaS [article]

Huaizheng Zhang, Yuanming Li, Yizheng Huang, Yonggang Wen, Jianxiong Yin, Kyle Guan
2020 arXiv   pre-print
MLModelCI provides multimedia researchers and developers with a one-stop platform for efficient machine learning (ML) services.  ...  In its essence, MLModelCI serves as a housekeeper to help users publish models.  ...  INTRODUCTION Machine Learning (ML) techniques, especially Deep Learning (DL), have been widely adopted into multimedia applications, ranging from video analysis to artwork generation.  ... 
arXiv:2006.05096v1 fatcat:ztg77t6vafdizg7cy6nhh72s3m

ModelHub.AI: Dissemination Platform for Deep Learning Models [article]

Ahmed Hosny, Michael Schwier, Christoph Berger, Evin P Örnek, Mehmet Turan, Phi V Tran, Leon Weninger, Fabian Isensee, Klaus H Maier-Hein, Richard McKinley, Michael T Lu, Udo Hoffmann (+4 others)
2019 arXiv   pre-print
learning models.  ...  Python and RESTful Application programming interfaces (APIs) enable users to interact with models hosted on ModelHub.AI and allows both researchers and developers to utilize models out-of-the-box.  ...  D ISCUSSION In this study, we present ModelHub, a publishing platform for the structured dissemination of pre-trained deep learning models.  ... 
arXiv:1911.13218v1 fatcat:4b4mdu4frzaozlgwaz6kvf4k7e

Open-Source Biomedical Image Analysis Models: A Meta-Analysis and Continuous Survey

Rui Li, Vaibhav Sharma, Subasini Thangamani, Artur Yakimovich
2022 Frontiers in Bioinformatics  
Recent advances in machine learning allow for unprecedented improvement in these approaches. However, these novel algorithms come with new requirements in order to remain open source.  ...  To understand how these requirements are met, we have collected 50 biomedical image analysis models and performed a meta-analysis of their respective papers, source code, dataset, and trained model parameters  ...  Data Science (DS), Machine Learning (ML), and Artificial Intelligence (AI) (Sonnenburg et al., 2007; Landset et al., 2015; Abadi et al., 2016; Paszke et al., 2019) .  ... 
doi:10.3389/fbinf.2022.912809 fatcat:wworclitpfeyxnt6hx6676azya

Parsl: Pervasive Parallel Programming in Python [article]

Yadu Babuji, Anna Woodard, Zhuozhao Li, Daniel S. Katz, Ben Clifford, Rohan Kumar, Lukasz Lacinski, Ryan Chard, Justin M. Wozniak, Ian Foster, Michael Wilde, Kyle Chard
2019 arXiv   pre-print
We show that these capabilities satisfy the needs of many-task, interactive, online, and machine learning applications in fields such as biology, cosmology, and materials science.  ...  Here, we present Parsl, a parallel scripting library that augments Python with simple, scalable, and flexible constructs for encoding parallelism.  ...  ACKNOWLEDGMENT This work was supported in part by NSF (ACI-1550588) and DOE (DE-AC02-06CH11357).  ... 
arXiv:1905.02158v1 fatcat:okcga7i4vza6zmx5lyj63seone

A Cloud-Based Framework for Machine Learning Workloads and Applications

Alvaro Lopez Garcia, Viet Tran, Andy S. Alic, Miguel Caballer, Isabel Campos Plasencia, Alessandro Costantini, Stefan Dlugolinsky, Doina Cristina Duma, Giacinto Donvito, Jorge Gomes, Ignacio Heredia Cacha, Jesus Marco De Lucas (+13 others)
In this paper we propose a distributed architecture to provide machine learning practitioners with a set of tools and cloud services that cover the whole machine learning development cycle: ranging from  ...  the models creation, training, validation and testing to the models serving as a service, sharing and publication.  ...  DLHub [29] puts the focus on publishing, sharing, and reusing machine learning models and their associated data, capturing its provenance and providing credit to authors.  ... 
doi:10.5445/ir/1000117464 fatcat:nuyifjwxmjb2zirh447fwiusqe

TomoGAN: Low-Dose Synchrotron X-Ray Tomography with Generative Adversarial Networks [article]

Zhengchun Liu, Tekin Bicer, Rajkumar Kettimuthu, Doga Gursoy, Francesco De Carlo, Ian Foster
2019 arXiv   pre-print
Evaluation with both simulated and experimental datasets shows that our approach can significantly reduce noise in reconstructed images, improving the structural similarity score of simulation and experimental  ...  Furthermore, the quality of the reconstructed images with filtered back projection followed by our denoising approach exceeds that of reconstructions with the simultaneous iterative reconstruction technique  ...  We gratefully acknowledge the computing resources provided and operated by the Joint Laboratory for System Evaluation at Argonne National Laboratory.  ... 
arXiv:1902.07582v4 fatcat:wgk23emajbbidc4hwyivqaog2a

Performance, Successes and Limitations of Deep Learning Semantic Segmentation of Multiple Defects in Transmission Electron Micrographs [article]

Ryan Jacobs, Mingren Shen, Yuhan Liu, Wei Hao, Xiaoshan Li, Ruoyu He, Jacob RC Greaves, Donglin Wang, Zeming Xie, Zitong Huang, Chao Wang, Kevin G. Field (+1 others)
2021 arXiv   pre-print
Overall, we find that the current model is a fast, effective tool for automatically characterizing and quantifying multiple defect types in microscopy images, with a level of accuracy on par with human  ...  We conduct an in-depth analysis of key model performance statistics, with a focus on quantities such as predicted distributions of defect shapes, defect sizes, and defect areal densities relevant to informing  ...  This work used the Extreme Science and Engineering Discovery Environment (XSEDE), which is supported by National Science  ... 
arXiv:2110.08244v1 fatcat:s7qaovfr6jf5vkqdkolmp5cbra

funcX: A Federated Function Serving Fabric for Science [article]

Ryan Chard, Yadu Babuji, Zhuozhao Li, Tyler Skluzacek, Anna Woodard, Ben Blaiszik, Ian Foster, Kyle Chard
2020 pre-print
can transform existing clouds, clusters, and supercomputers into function serving systems, while funcX's cloud-hosted service provides transparent, secure, and reliable function execution across a federated  ...  We motivate the need for funcX with several scientific case studies, present our prototype design and implementation, show optimizations that deliver throughput in excess of 1 million functions per second  ...  Department of Energy under Contract DE-AC02-06CH11357 and used resources of the Argonne Leadership Computing Facility.  ... 
doi:10.1145/3369583.3392683 arXiv:2005.04215v1 fatcat:stvzh7eslnf65e3nzbcvl5aizu

Serverless Computing (Dagstuhl Seminar 21201)

Cristina Abad, Ian T. Foster, Nikolas Herbst, Alexandru Iosup
Several related but disjoint fields, notably software and systems engineering, parallel and distributed systems, and system and performance analysis and modeling, aim to address these opportunities and  ...  and academia.  ...  with a high degree of reusability for distributed data sources.  ... 
doi:10.4230/dagrep.11.4.34 fatcat:bkqp4ttlv5hu5flhxifw6piaom

From Platform to Knowledge Graph: Evolution of Laboratory Automation

Jiaru Bai, Liwei Cao, Sebastian Mosbach, Jethro Akroyd, Alexei Lapkin, Markus Kraft, Apollo-University Of Cambridge Repository
High-fidelity computer-aided experimentation is becoming more accessible with the development of computing power and artificial intelligence tools.  ...  Emerging technologies for interoperable data representation and multi-agent systems are also discussed with their recent applications in chemical automation.  ...  A relevant first step toward addressing this issue is demonstrated by DLHub, 152 which allows users to publish, share, and cite ML models for applications in science.  ... 
doi:10.17863/cam.83288 fatcat:igsnqxp52fhlxowgcbck7ub2xu