The challenges of TinyML implementation: A literature review
Loading...
Supplementary material
Other Title
Authors
Adlakha, Riya
Kabbar, Eltahir
Kabbar, Eltahir
Author ORCID Profiles (clickable)
Degree
Grantor
Date
2024-07-24
Supervisors
Type
Conference Contribution - Paper in Published Proceedings
Ngā Upoko Tukutuku (Māori subject headings)
Keyword
TinyML
machine learning
microcontrollers
Internet of Things (IoT)
edge computing
literature reviews
machine learning
microcontrollers
Internet of Things (IoT)
edge computing
literature reviews
ANZSRC Field of Research Code (2020)
Citation
Adlakha, R., & Kabbar, E. (2024). The challenges of TinyML implementation: A literature review. In H. Sharifzadeh (Ed.), Proceedings: CITRENZ 2023 Conference, Auckland, 27–29 September (pp. 160–169). ePress, Unitec. Auckland, New Zealand. https://doi.org/10.34074/proc.240120
Abstract
This study aims to sensitise and summarise the tiny machine learning (TinyML) implementation literature. TinyML is a subset of machine learning (ML) that focuses on implementing ML models on resource constrained devices such as microcontrollers, embedded systems, and internet of things (IoT) devices. A systematic literature review is performed on the works published in this field in the last decade. The key focus of this article is to understand the critical challenges faced by this emerging technology. We present five significant challenges of TinyML, namely, limited and dynamic resources, heterogeneity, network management, security and privacy, and model design. This article will be of interest to researchers and practitioners who are interested in the fields of ML, IoT and edge computing.
Publisher
Unitec ePress, Te Pūkenga - New Zealand Institute of Skills and Technology
Permanent link
Link to ePress publication
DOI
https://doi.org/10.34074/proc.240120
Copyright holder
Authors
Copyright notice
CC BY-NC Attribution-NonCommercial 4.0 International