Work-in-progress: BPNet: Branch-pruned conditional neural network for systematic time-accuracy tradeoff in DNN inference

Kyungchul Park, Youngmin Yi

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Scopus citations

Abstract

Recently, there have been attempts to execute the neural network conditionally with auxiliary classifiers allowing early termination depending on the difficulty of the input, which can reduce the execution time or energy consumption without any or with negligible accuracy decrease. However, these studies do not consider how many or where the auxiliary classifiers, or branches, should be added in a systematic fashion. In this paper, we propose Branch-pruned Conditional Neural Network (BPNet) and its methodology in which the time-accuracy tradeoff for the conditional neural network can be found systematically. We applied BPNet to SqueezeNet, ResNet-20, and VGG-16 with CIFAR-10 and 100. BPNet achieves on average 2.0× of speedups without any accuracy drop on average compared to the base network.

Original languageEnglish
Title of host publicationProceedings of the International Conference on Hardware/Software Codesign and System Synthesis Companion, CODES/ISSS 2019
PublisherAssociation for Computing Machinery, Inc
ISBN (Electronic)9781450369237
DOIs
StatePublished - 13 Oct 2019
Event2019 International Conference on Hardware/Software Codesign and System Synthesis, CODES/ISSS 2019 - New York, United States
Duration: 13 Oct 201918 Oct 2019

Publication series

NameProceedings of the International Conference on Hardware/Software Codesign and System Synthesis Companion, CODES/ISSS 2019

Conference

Conference2019 International Conference on Hardware/Software Codesign and System Synthesis, CODES/ISSS 2019
Country/TerritoryUnited States
CityNew York
Period13/10/1918/10/19

Fingerprint

Dive into the research topics of 'Work-in-progress: BPNet: Branch-pruned conditional neural network for systematic time-accuracy tradeoff in DNN inference'. Together they form a unique fingerprint.

Cite this