CANDLE/Supervisor: A workflow framework for machine learning applied to cancer research

Justin M. Wozniak; Rajeev Jain; Prasanna Balaprakash; Jonathan Ozik; Nicholson T. Collier; John Bauer; Fangfang Xia; Thomas Brettin; Rick Stevens; Jamaludin Mohd-Yusof; Cristina Garcia Cardona; Brian Van Essen; Matthew Baughman

doi:10.1186/s12859-018-2508-4

CANDLE/Supervisor: A workflow framework for machine learning applied to cancer research

Justin M. Wozniak, Rajeev Jain, Prasanna Balaprakash, Jonathan Ozik, Nicholson T. Collier, John Bauer, Fangfang Xia, Thomas Brettin, Rick Stevens, Jamaludin Mohd-Yusof, Cristina Garcia Cardona, Brian Van Essen, Matthew Baughman

Research output: Contribution to journal › Article › peer-review

55 Scopus citations

Abstract

Background: Current multi-petaflop supercomputers are powerful systems, but present challenges when faced with problems requiring large machine learning workflows. Complex algorithms running at system scale, often with different patterns that require disparate software packages and complex data flows cause difficulties in assembling and managing large experiments on these machines. Results: This paper presents a workflow system that makes progress on scaling machine learning ensembles, specifically in this first release, ensembles of deep neural networks that address problems in cancer research across the atomistic, molecular and population scales. The initial release of the application framework that we call CANDLE/Supervisor addresses the problem of hyper-parameter exploration of deep neural networks. Conclusions: Initial results demonstrating CANDLE on DOE systems at ORNL, ANL and NERSC (Titan, Theta and Cori, respectively) demonstrate both scaling and multi-platform execution.

Original language	English
Article number	491
Journal	BMC Bioinformatics
Volume	19
DOIs	https://doi.org/10.1186/s12859-018-2508-4
State	Published - Dec 21 2018
Externally published	Yes

Funding

This material is based upon work supported by the U.S. Department of Energy, Office of Science, under contract number DE-AC02-06CH11357. This research was supported by the Exascale Computing Project (17-SC-20-SC), a collaborative effort of the U.S. Department of Energy Office of Science and the National Nuclear Security Administration. This research used resources of the Argonne Leadership Computing Facility, which is a DOE Office of Science User Facility. This research used resources of the Argonne Leadership Computing Facility, which is a DOE Office of Science User Facility supported under Contract DE-AC02-06CH11357. This research used resources of the Oak Ridge Leadership Computing Facility at the Oak Ridge National Laboratory, which is supported by the Office of Science of the U.S. Department of Energy under Contract No. DE-AC05-00OR22725. This research used resources of the National Energy Research Scientific Computing Center, a DOE Office of Science User Facility supported by the Office of Science of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231. This material is based upon work supported by the NIH (R01GM115839). Publication costs were funded by the CANDLE project under the Exascale Computing Project, a U.S. Department of Energy program.

Funders	Funder number
DOE Office of Science
U.S. Department of Energy Office of Science
National Institutes of Health
U.S. Department of Energy
National Institute of General Medical Sciences	R01GM115839
Office of Science	17-SC-20-SC, DE-AC02-06CH11357
National Nuclear Security Administration

Keywords

Article
Author
Sample

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1186/s12859-018-2508-4

Cite this

@article{a59cd9138c264596befff5f06ecb5d75,

title = "CANDLE/Supervisor: A workflow framework for machine learning applied to cancer research",

abstract = "Background: Current multi-petaflop supercomputers are powerful systems, but present challenges when faced with problems requiring large machine learning workflows. Complex algorithms running at system scale, often with different patterns that require disparate software packages and complex data flows cause difficulties in assembling and managing large experiments on these machines. Results: This paper presents a workflow system that makes progress on scaling machine learning ensembles, specifically in this first release, ensembles of deep neural networks that address problems in cancer research across the atomistic, molecular and population scales. The initial release of the application framework that we call CANDLE/Supervisor addresses the problem of hyper-parameter exploration of deep neural networks. Conclusions: Initial results demonstrating CANDLE on DOE systems at ORNL, ANL and NERSC (Titan, Theta and Cori, respectively) demonstrate both scaling and multi-platform execution.",

keywords = "Article, Author, Sample",

author = "Wozniak, {Justin M.} and Rajeev Jain and Prasanna Balaprakash and Jonathan Ozik and Collier, {Nicholson T.} and John Bauer and Fangfang Xia and Thomas Brettin and Rick Stevens and Jamaludin Mohd-Yusof and Cardona, {Cristina Garcia} and Essen, {Brian Van} and Matthew Baughman",

note = "Publisher Copyright: {\textcopyright} 2018 The Author(s).",

year = "2018",

month = dec,

day = "21",

doi = "10.1186/s12859-018-2508-4",

language = "English",

volume = "19",

journal = "BMC Bioinformatics",

issn = "1471-2105",

publisher = "BioMed Central",

}

TY - JOUR

T1 - CANDLE/Supervisor

T2 - A workflow framework for machine learning applied to cancer research

AU - Wozniak, Justin M.

AU - Jain, Rajeev

AU - Balaprakash, Prasanna

AU - Ozik, Jonathan

AU - Collier, Nicholson T.

AU - Bauer, John

AU - Xia, Fangfang

AU - Brettin, Thomas

AU - Stevens, Rick

AU - Mohd-Yusof, Jamaludin

AU - Cardona, Cristina Garcia

AU - Essen, Brian Van

AU - Baughman, Matthew

PY - 2018/12/21

Y1 - 2018/12/21

N2 - Background: Current multi-petaflop supercomputers are powerful systems, but present challenges when faced with problems requiring large machine learning workflows. Complex algorithms running at system scale, often with different patterns that require disparate software packages and complex data flows cause difficulties in assembling and managing large experiments on these machines. Results: This paper presents a workflow system that makes progress on scaling machine learning ensembles, specifically in this first release, ensembles of deep neural networks that address problems in cancer research across the atomistic, molecular and population scales. The initial release of the application framework that we call CANDLE/Supervisor addresses the problem of hyper-parameter exploration of deep neural networks. Conclusions: Initial results demonstrating CANDLE on DOE systems at ORNL, ANL and NERSC (Titan, Theta and Cori, respectively) demonstrate both scaling and multi-platform execution.

AB - Background: Current multi-petaflop supercomputers are powerful systems, but present challenges when faced with problems requiring large machine learning workflows. Complex algorithms running at system scale, often with different patterns that require disparate software packages and complex data flows cause difficulties in assembling and managing large experiments on these machines. Results: This paper presents a workflow system that makes progress on scaling machine learning ensembles, specifically in this first release, ensembles of deep neural networks that address problems in cancer research across the atomistic, molecular and population scales. The initial release of the application framework that we call CANDLE/Supervisor addresses the problem of hyper-parameter exploration of deep neural networks. Conclusions: Initial results demonstrating CANDLE on DOE systems at ORNL, ANL and NERSC (Titan, Theta and Cori, respectively) demonstrate both scaling and multi-platform execution.

KW - Article

KW - Author

KW - Sample

UR - http://www.scopus.com/inward/record.url?scp=85058909711&partnerID=8YFLogxK

U2 - 10.1186/s12859-018-2508-4

DO - 10.1186/s12859-018-2508-4

M3 - Article

C2 - 30577736

AN - SCOPUS:85058909711

SN - 1471-2105

VL - 19

JO - BMC Bioinformatics

JF - BMC Bioinformatics

M1 - 491

ER -

CANDLE/Supervisor: A workflow framework for machine learning applied to cancer research

Abstract

Funding

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this