Dynamic Image Difficulty-Aware DNN Pruning

Deep Neural Networks (DNNs) have achieved impressive performance in various image recognition tasks, but their large model sizes make them challenging to deploy on resource-constrained devices. In this paper, we propose a dynamic DNN pruning approach that takes into account the difficulty of the inc...

Full description

Saved in:
Bibliographic Details
Published inMicromachines (Basel) Vol. 14; no. 5; p. 908
Main Authors Pentsos, Vasileios, Spantidi, Ourania, Anagnostopoulos, Iraklis
Format Journal Article
LanguageEnglish
Published Switzerland MDPI AG 23.04.2023
MDPI
Subjects
Accuracy
Artificial neural networks
Classification
Datasets
Deep Neural Networks
Embedded systems
Energy consumption
Human subjects
image difficulty
Neural networks
Pruning
Sparsity
Support vector machines
pruning
image difficulty
Deep Neural Networks
embedded systems
Online AccessGet full text

Cover

More Information
Summary:Deep Neural Networks (DNNs) have achieved impressive performance in various image recognition tasks, but their large model sizes make them challenging to deploy on resource-constrained devices. In this paper, we propose a dynamic DNN pruning approach that takes into account the difficulty of the incoming images during inference. To evaluate the effectiveness of our method, we conducted experiments on the ImageNet dataset on several state-of-art DNNs. Our results show that the proposed approach reduces the model size and amount of DNN operations without the need to retrain or fine-tune the pruned model. Overall, our method provides a promising direction for designing efficient frameworks for lightweight DNN models that can adapt to the varying complexity of input images.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
These authors contributed equally to this work.
ISSN:2072-666X
2072-666X
DOI:10.3390/mi14050908