Data Aware Differentiable Neural Architecture Search for Tiny Keyword Spotting Applications

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Shi, Yujia, Njor, Emil, Martínez-Nuevo, Pablo, Shepstone, Sven Ewan, Fafoutis, Xenofon
Format:	Preprint
Published:	2025
Subjects:	Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

_version_	1866918099894140928
author	Shi, Yujia Njor, Emil Martínez-Nuevo, Pablo Shepstone, Sven Ewan Fafoutis, Xenofon
author_facet	Shi, Yujia Njor, Emil Martínez-Nuevo, Pablo Shepstone, Sven Ewan Fafoutis, Xenofon
contents	The success of Machine Learning is increasingly tempered by its significant resource footprint, driving interest in efficient paradigms like TinyML. However, the inherent complexity of designing TinyML systems hampers their broad adoption. To reduce this complexity, we introduce "Data Aware Differentiable Neural Architecture Search". Unlike conventional Differentiable Neural Architecture Search, our approach expands the search space to include data configuration parameters alongside architectural choices. This enables Data Aware Differentiable Neural Architecture Search to co-optimize model architecture and input data characteristics, effectively balancing resource usage and system performance for TinyML applications. Initial results on keyword spotting demonstrate that this novel approach to TinyML system design can generate lean but highly accurate systems.
format	Preprint
id	arxiv_https___arxiv_org_abs_2507_15545
institution	arXiv
publishDate	2025
record_format	arxiv
spellingShingle	Data Aware Differentiable Neural Architecture Search for Tiny Keyword Spotting Applications Shi, Yujia Njor, Emil Martínez-Nuevo, Pablo Shepstone, Sven Ewan Fafoutis, Xenofon Machine Learning The success of Machine Learning is increasingly tempered by its significant resource footprint, driving interest in efficient paradigms like TinyML. However, the inherent complexity of designing TinyML systems hampers their broad adoption. To reduce this complexity, we introduce "Data Aware Differentiable Neural Architecture Search". Unlike conventional Differentiable Neural Architecture Search, our approach expands the search space to include data configuration parameters alongside architectural choices. This enables Data Aware Differentiable Neural Architecture Search to co-optimize model architecture and input data characteristics, effectively balancing resource usage and system performance for TinyML applications. Initial results on keyword spotting demonstrate that this novel approach to TinyML system design can generate lean but highly accurate systems.
title	Data Aware Differentiable Neural Architecture Search for Tiny Keyword Spotting Applications
topic	Machine Learning
url	https://arxiv.org/abs/2507.15545