Optimising resource management for embedded machine learning
Optimising resource management for embedded machine learning
Machine learning inference is increasingly being executed locally on mobile and embedded platforms, due to the clear advantages in latency, privacy and connectivity. In this paper, we present approaches for online resource management in heterogeneous multi-core systems and show how they can be applied to optimise the performance of machine learning workloads. Performance can be defined using platform-dependent (e.g. speed, energy) and platform-independent (accuracy, confidence) metrics. In particular, we show how a Deep Neural Network (DNN) can be dynamically scalable to trade-off these various performance metrics. Achieving consistent performance when executing on different platforms is necessary yet challenging, due to the different resources provided and their capability, and their time-varying availability when executing alongside other workloads. Managing the interface between available hardware resources (often numerous and heterogeneous in nature), software requirements, and user experience is increasingly complex.
Dynamic Deep Neural Network, Embedded Machine Learning, Runtime Resource Management
1556-1561
Xun, Lei
51a0da82-6979-49a8-8eff-ada011f5aff5
Tran-Thanh, Long
e0666669-d34b-460e-950d-e8b139fab16c
Al-Hashimi, Bashir
0b29c671-a6d2-459c-af68-c4614dce3b5d
Merrett, Geoff
89b3a696-41de-44c3-89aa-b0aa29f54020
March 2020
Xun, Lei
51a0da82-6979-49a8-8eff-ada011f5aff5
Tran-Thanh, Long
e0666669-d34b-460e-950d-e8b139fab16c
Al-Hashimi, Bashir
0b29c671-a6d2-459c-af68-c4614dce3b5d
Merrett, Geoff
89b3a696-41de-44c3-89aa-b0aa29f54020
Xun, Lei, Tran-Thanh, Long, Al-Hashimi, Bashir and Merrett, Geoff
(2020)
Optimising resource management for embedded machine learning.
Di Natale, Giorgio, Bolchini, Cristiana and Vatajelu, Elena-Ioana
(eds.)
In Proceedings of the 2020 Design, Automation and Test in Europe Conference and Exhibition, DATE 2020.
.
(doi:10.23919/DATE48585.2020.9116235).
Record type:
Conference or Workshop Item
(Paper)
Abstract
Machine learning inference is increasingly being executed locally on mobile and embedded platforms, due to the clear advantages in latency, privacy and connectivity. In this paper, we present approaches for online resource management in heterogeneous multi-core systems and show how they can be applied to optimise the performance of machine learning workloads. Performance can be defined using platform-dependent (e.g. speed, energy) and platform-independent (accuracy, confidence) metrics. In particular, we show how a Deep Neural Network (DNN) can be dynamically scalable to trade-off these various performance metrics. Achieving consistent performance when executing on different platforms is necessary yet challenging, due to the different resources provided and their capability, and their time-varying availability when executing alongside other workloads. Managing the interface between available hardware resources (often numerous and heterogeneous in nature), software requirements, and user experience is increasingly complex.
Text
Optimising resource management for embedded machine learning
- Version of Record
Text
Optimising resource management for embedded machine learning_v2
- Other
Text
Optimising resource management for embedded machine learning_v3
- Other
More information
Accepted/In Press date: 30 October 2019
Published date: March 2020
Additional Information:
Funding Information:
This work was supported in part by the Engineering and Physical Sciences Research Council (EPSRC) under Grant EP/S030069/1. Experimental data can be found at www. eprints.soton.ac.uk
Publisher Copyright:
© 2020 EDAA.
Copyright:
Copyright 2020 Elsevier B.V., All rights reserved.
Keywords:
Dynamic Deep Neural Network, Embedded Machine Learning, Runtime Resource Management
Identifiers
Local EPrints ID: 436228
URI: http://eprints.soton.ac.uk/id/eprint/436228
PURE UUID: 3525323f-b950-4620-a008-2f0bb3612852
Catalogue record
Date deposited: 04 Dec 2019 17:30
Last modified: 17 Mar 2024 03:02
Export record
Altmetrics
Contributors
Author:
Lei Xun
Author:
Long Tran-Thanh
Author:
Bashir Al-Hashimi
Author:
Geoff Merrett
Editor:
Giorgio Di Natale
Editor:
Cristiana Bolchini
Editor:
Elena-Ioana Vatajelu
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics