Article (Scientific journals)
CalcGraph: taming the high costs of deep learning using models
Lorentz, Joe; Hartmann, Thomas; Moawad, Assaad et al.
2022In Software and Systems Modeling
Peer Reviewed verified by ORBi
 

Files


Full Text
accepted_version_CalcGraph: taming the high costs of deep learning using models.pdf
Author postprint (758.77 kB)
Download

This version of the article has been accepted for publication, after peer review but is not the Version of Record and does not reflect post-acceptance improvements, or any corrections. The Version of Record is available online at: https://doi.org/10.1007/s10270-022-01052-7


All documents in ORBilu are protected by a user license.

Send to



Details



Keywords :
Differentiable programming; Computational graph model; Edge AI
Abstract :
[en] Models based on differential programming, like deep neural networks, are well established in research and able to outperform manually coded counterparts in many applications. Today, there is a rising interest to introduce this flexible modeling to solve real-world problems. A major challenge when moving from research to application is the strict constraints on computational resources (memory and time). It is difficult to determine and contain the resource requirements of differential models, especially during the early training and hyperparameter exploration stages. In this article, we address this challenge by introducing CalcGraph, a model abstraction of differentiable programming layers. CalcGraph allows to model the computational resources that should be used and then CalcGraph’s model interpreter can automatically schedule the execution respecting the specifications made. We propose a novel way to efficiently switch models from storage to preallocated memory zones and vice versa to maximize the number of model executions given the available resources. We demonstrate the efficiency of our approach by showing that it consumes less resources than state-of-the-art frameworks like TensorFlow and PyTorch for single-model and multi-model execution.
Disciplines :
Computer science
Author, co-author :
Lorentz, Joe ;  DataThings S.A. ; University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT)
Hartmann, Thomas ;  DataThings S.A.
Moawad, Assaad ;  DataThings S.A.
Fouquet, François ;  DataThings S.A.
Aouada, Djamila  ;  University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > CVI2
Le Traon, Yves ;  University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > SerVal
External co-authors :
no
Language :
English
Title :
CalcGraph: taming the high costs of deep learning using models
Publication date :
25 October 2022
Journal title :
Software and Systems Modeling
ISSN :
1619-1374
Publisher :
Springer, Germany
Peer reviewed :
Peer Reviewed verified by ORBi
Focus Area :
Computational Sciences
FnR Project :
FNR14297122 - Towards Edge-optimized Deep Learning For Explainable Quality Control, 2019 (01/01/2020-31/12/2023) - Joe Lorentz
Available on ORBilu :
since 24 November 2022

Statistics


Number of views
80 (4 by Unilu)
Number of downloads
65 (2 by Unilu)

Scopus citations®
 
0
Scopus citations®
without self-citations
0
OpenCitations
 
0
WoS citations
 
0

Bibliography


Similar publications



Contact ORBilu