CaW-NAS: Compression Aware Neural Architecture Search

With the ever-growing demand for deep learning (DL) at the edge, building small and efficient DL architectures has become a significant challenge. Optimization techniques such as quantization, pruning or hardware-aware neural architecture search (HW-NAS) have been proposed. In this paper, we present...

Full description

Saved in:
Bibliographic Details
Published in2022 25th Euromicro Conference on Digital System Design (DSD) pp. 391 - 397
Main Authors Benmeziane, Hadjer, Ouranoughi, Hamza, Niar, Smail, El Maghraoui, Kaoutar
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.08.2022
Subjects
Architecture
Buildings
computer vision
Deep learning
Digital systems
Energy consumption
hardware-aware neural architecture search
quantization
Quantization (signal)
Search problems
Online AccessGet full text

Cover

More Information
Summary:With the ever-growing demand for deep learning (DL) at the edge, building small and efficient DL architectures has become a significant challenge. Optimization techniques such as quantization, pruning or hardware-aware neural architecture search (HW-NAS) have been proposed. In this paper, we present an efficient HW-NAS; Compression-Aware Neural Architecture search (CaW-NAS), that combines the search for the architecture and its quantization policy. While former works search over a fully quantized search space, we define our search space with quantized and non-quantized architectures. Our search strategy finds the best trade-off between accuracy and latency according to the target hardware. Experimental results on a mobile platform show that, our method allows to obtain more efficient networks in terms of accuracy, execution time and energy consumption when compared to the state of the art.
ISSN:2771-2508
DOI:10.1109/DSD57027.2022.00059