The University of Southampton
University of Southampton Institutional Repository

Adapting branched networks to realise progressive intelligence

Adapting branched networks to realise progressive intelligence
Adapting branched networks to realise progressive intelligence
Progressive intelligence is a formulation of machine learning which trades-off performance requirements with resource availability. It does this by approaching the inference process incrementally. Current work in this area focuses on overall model performance rather than optimising its complete operating range. In this paper, we build upon existing explainability and branched neural network research to show how neural networks can be adapted to exhibit progressive intelligence. We assess the utility of joint branch optimisation for progressive intelligence using a number of explainability metrics. When optimising the area under curve of layerwise linear probe accuracy we find equally weighted early-exit branch optimisation produces models with the highest linear probe accuracy throughout the backbone. By varying confidence thresholds we represent the entire range over which the model can operate, we then explore its interaction with the scaling of the branched neural network backbone. Finally, we propose a novel ensemble inference strategy which utilises repeat predictions and requires no additional optimisation. Experiments with CIFAR10/100 show that this inference strategy can save up to 44% of the multiply accumulate operations used in inference whilst maintaining model performance, when compared against conventional early-exit methods.
Dymond, Jack
3309494f-9bac-4891-a6a7-06a900eebf87
Stein, Sebastian
cb2325e7-5e63-475e-8a69-9db2dfbdb00b
Gunn, Stephen
306af9b3-a7fa-4381-baf9-5d6a6ec89868
Dymond, Jack
3309494f-9bac-4891-a6a7-06a900eebf87
Stein, Sebastian
cb2325e7-5e63-475e-8a69-9db2dfbdb00b
Gunn, Stephen
306af9b3-a7fa-4381-baf9-5d6a6ec89868

Dymond, Jack, Stein, Sebastian and Gunn, Stephen (2022) Adapting branched networks to realise progressive intelligence. British Machine Vision Conference, , London, United Kingdom. 21 - 24 Nov 2022. 13 pp .

Record type: Conference or Workshop Item (Paper)

Abstract

Progressive intelligence is a formulation of machine learning which trades-off performance requirements with resource availability. It does this by approaching the inference process incrementally. Current work in this area focuses on overall model performance rather than optimising its complete operating range. In this paper, we build upon existing explainability and branched neural network research to show how neural networks can be adapted to exhibit progressive intelligence. We assess the utility of joint branch optimisation for progressive intelligence using a number of explainability metrics. When optimising the area under curve of layerwise linear probe accuracy we find equally weighted early-exit branch optimisation produces models with the highest linear probe accuracy throughout the backbone. By varying confidence thresholds we represent the entire range over which the model can operate, we then explore its interaction with the scaling of the branched neural network backbone. Finally, we propose a novel ensemble inference strategy which utilises repeat predictions and requires no additional optimisation. Experiments with CIFAR10/100 show that this inference strategy can save up to 44% of the multiply accumulate operations used in inference whilst maintaining model performance, when compared against conventional early-exit methods.

Text
Adapting branched networks to realise progressive intelligence - Accepted Manuscript
Available under License Other.
Download (379kB)

More information

Accepted/In Press date: September 2022
e-pub ahead of print date: 22 November 2022
Venue - Dates: British Machine Vision Conference, , London, United Kingdom, 2022-11-21 - 2022-11-24

Identifiers

Local EPrints ID: 471457
URI: http://eprints.soton.ac.uk/id/eprint/471457
PURE UUID: e47f7314-d0be-4c74-aae0-24732d78ccd0
ORCID for Sebastian Stein: ORCID iD orcid.org/0000-0003-2858-8857

Catalogue record

Date deposited: 08 Nov 2022 18:34
Last modified: 11 Nov 2022 02:39

Export record

Contributors

Author: Jack Dymond
Author: Sebastian Stein ORCID iD
Author: Stephen Gunn

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

Atom RSS 1.0 RSS 2.0

Contact ePrints Soton: eprints@soton.ac.uk

ePrints Soton supports OAI 2.0 with a base URL of http://eprints.soton.ac.uk/cgi/oai2

This repository has been built using EPrints software, developed at the University of Southampton, but available to everyone to use.

We use cookies to ensure that we give you the best experience on our website. If you continue without changing your settings, we will assume that you are happy to receive cookies on the University of Southampton website.

×