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

Autor: Wozniak JM; Argonne National Laboratory, Argonne, IL, USA. woz@anl.gov., Jain R; Argonne National Laboratory, Argonne, IL, USA., Balaprakash P; Argonne National Laboratory, Argonne, IL, USA., Ozik J; Argonne National Laboratory, Argonne, IL, USA., Collier NT; Argonne National Laboratory, Argonne, IL, USA., Bauer J; Argonne National Laboratory, Argonne, IL, USA., Xia F; Argonne National Laboratory, Argonne, IL, USA., Brettin T; Argonne National Laboratory, Argonne, IL, USA., Stevens R; Argonne National Laboratory, Argonne, IL, USA., Mohd-Yusof J; Los Alamos National Laboratory, Los Alamos, NM, USA., Cardona CG; Los Alamos National Laboratory, Los Alamos, NM, USA., Essen BV; Lawrence Livermore National Laboratory, Livermore, CA, USA., Baughman M; Minerva, San Francisco, CA, USA.
Jazyk: angličtina
Zdroj: BMC bioinformatics [BMC Bioinformatics] 2018 Dec 21; Vol. 19 (Suppl 18), pp. 491. Date of Electronic Publication: 2018 Dec 21.
DOI: 10.1186/s12859-018-2508-4
Abstrakt: 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.
Databáze: MEDLINE
Externí odkaz:
Nepřihlášeným uživatelům se plný text nezobrazuje