Publication Type
Conference Paper
Book Title
Proceedings of the 2023 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW)
Publication Date
Page Numbers
795 to 798
Publisher Location
New Jersey, United States of America
Conference Name
Scalable Deep Learning over Parallel And Distributed Infrastructures
Conference Location
St. Petersburg, Florida, United States of America
Conference Sponsor
IPDPS
Conference Date
Abstract
As part of a larger effort, this work-in-progress reports the possible advantages of modifying conventional workflows used to generate labelled training samples and train machine learning (ML) models on them. We compare results from three different workflows using neutron scattering data analysis as the motivating application and report about 20% improvement in speedup, with no appreciable loss of model accuracy, over a baseline workflow.
