Lightweight machine learning-enabled supervised learning algorithms for resource-constrained consumer electronics at the edge

doi:10.1016/j.array.2026.100878

Lightweight machine learning-enabled supervised learning algorithms for resource-constrained consumer electronics at the edge

2026-05-05

0

OA

AI

A

Abdullah Ayub Khan *

A

Abdul Khalique Shaikh *

H

Hamad Almansour

J

Jamil Abedalrahim Jamil Alsayaydeh *

K

Kaamran Raahemifar

M

Mohamad Afendee Mohamed

DOI：10.1016/j.array.2026.100878

Original

Save

View PDF

Abstract

En 中文

• Due to resource constraints, the design of traditional machine learning models has become more difficult as a result of the dramatic shift towards edge computing technology, particularly the move away from electronics consumers brought about by the growth of the Internet of Things and intelligent devices. The primary driving force behind this paper is the potential to close this gap by examining and introducing lightweight machine learning methods in addition to optimization strategies like model quantization and pruning, particularly for edge computing environments. • In this research, we report on an investigation that shows how well several lightweight machine learning algorithms perform in resource-constrained environments. • We demonstrate their efficacy using case studies in anomaly detection, picture classification, and voice recognition. • This paper helps developers advance the field of edge-based machine learning for next-generation consumer electronics, this paper offers helpful advice on algorithm selection and optimization for edge devices.

Keywords:

Artificial intelligence (AI)

Machine learning (ML)

Internet of things (IoT)

Supervised learning

Edge computing

Lightweight machine learning-enabled supervised learning algorithms for resource-constrained consumer electronics at the edge

Abstract

AI Summary

Journal

Organization