Enhancing Early Exit Performance With Uncertainty-Aware Training in Convolutional Neural Networks for Image Classification

Deep Neural Networks have become more and more complex over the past years delivering efficient results but demanding more requirements for resources. The traditional cloud-based DNN inference produces poor real-time performance due to its high latency requirement in the network. Nowadays, fog and e...

Full description

Saved in:
Bibliographic Details
Main Authors: Thanu Kurian, Somasundaram Thangam
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11008656/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849471963719794688
author Thanu Kurian
Somasundaram Thangam
author_facet Thanu Kurian
Somasundaram Thangam
author_sort Thanu Kurian
collection DOAJ
description Deep Neural Networks have become more and more complex over the past years delivering efficient results but demanding more requirements for resources. The traditional cloud-based DNN inference produces poor real-time performance due to its high latency requirement in the network. Nowadays, fog and edge have been popular, where computation occurs near the data source or between the source and the cloud. High usage of computational resources by DNN incurs limited room for its deployment in such resource-constrained devices. Deep Neural Networks with multiple exit architecture are an optimal approach to preserve time and computational resources by predicting results at an early stage using multiple early exits to support Edge and Fog Intelligence. Early exit, although lightweight and energy efficient, mostly exhibits degraded performance compared to later exits due to the count of samples that are misclassified. To overcome the downfall, we propose uncertainty estimation in the loss function as an additional parameter for classification. Observational results on CIFAR, MNIST, human lung CT scan, and brain MRI datasets using VGG-16, Lenet-5, and Resnet architectures exhibit that our proposed model is better than the existing models to improve performance as well as prediction confidence in Deep Learning Models with early exits.
format Article
id doaj-art-37a6e77b7ab84a3fba65e3be7cdb3c7c
institution Kabale University
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-37a6e77b7ab84a3fba65e3be7cdb3c7c2025-08-20T03:24:39ZengIEEEIEEE Access2169-35362025-01-0113916719168110.1109/ACCESS.2025.357241511008656Enhancing Early Exit Performance With Uncertainty-Aware Training in Convolutional Neural Networks for Image ClassificationThanu Kurian0https://orcid.org/0009-0006-1559-2645Somasundaram Thangam1https://orcid.org/0000-0001-9284-7724Department of Computer Science and Engineering, Amrita School of Computing, Amrita Vishwa Vidyapeetham, Bengaluru, IndiaDepartment of Computer Science and Engineering, Amrita School of Computing, Amrita Vishwa Vidyapeetham, Bengaluru, IndiaDeep Neural Networks have become more and more complex over the past years delivering efficient results but demanding more requirements for resources. The traditional cloud-based DNN inference produces poor real-time performance due to its high latency requirement in the network. Nowadays, fog and edge have been popular, where computation occurs near the data source or between the source and the cloud. High usage of computational resources by DNN incurs limited room for its deployment in such resource-constrained devices. Deep Neural Networks with multiple exit architecture are an optimal approach to preserve time and computational resources by predicting results at an early stage using multiple early exits to support Edge and Fog Intelligence. Early exit, although lightweight and energy efficient, mostly exhibits degraded performance compared to later exits due to the count of samples that are misclassified. To overcome the downfall, we propose uncertainty estimation in the loss function as an additional parameter for classification. Observational results on CIFAR, MNIST, human lung CT scan, and brain MRI datasets using VGG-16, Lenet-5, and Resnet architectures exhibit that our proposed model is better than the existing models to improve performance as well as prediction confidence in Deep Learning Models with early exits.https://ieeexplore.ieee.org/document/11008656/Deep neural networksearly exitedge computingfog computingMonte Carlo dropoutuncertainty estimation
spellingShingle Thanu Kurian
Somasundaram Thangam
Enhancing Early Exit Performance With Uncertainty-Aware Training in Convolutional Neural Networks for Image Classification
IEEE Access
Deep neural networks
early exit
edge computing
fog computing
Monte Carlo dropout
uncertainty estimation
title Enhancing Early Exit Performance With Uncertainty-Aware Training in Convolutional Neural Networks for Image Classification
title_full Enhancing Early Exit Performance With Uncertainty-Aware Training in Convolutional Neural Networks for Image Classification
title_fullStr Enhancing Early Exit Performance With Uncertainty-Aware Training in Convolutional Neural Networks for Image Classification
title_full_unstemmed Enhancing Early Exit Performance With Uncertainty-Aware Training in Convolutional Neural Networks for Image Classification
title_short Enhancing Early Exit Performance With Uncertainty-Aware Training in Convolutional Neural Networks for Image Classification
title_sort enhancing early exit performance with uncertainty aware training in convolutional neural networks for image classification
topic Deep neural networks
early exit
edge computing
fog computing
Monte Carlo dropout
uncertainty estimation
url https://ieeexplore.ieee.org/document/11008656/
work_keys_str_mv AT thanukurian enhancingearlyexitperformancewithuncertaintyawaretraininginconvolutionalneuralnetworksforimageclassification
AT somasundaramthangam enhancingearlyexitperformancewithuncertaintyawaretraininginconvolutionalneuralnetworksforimageclassification