Single-layer KAN for deepfake classification: Balancing efficiency and performance in resource constrained environments.

Deepfakes, synthetic media created using artificial intelligence, threaten the authenticity of digital content. Traditional detection methods, such as Convolutional Neural Networks (CNNs), require substantial computational resources, rendering them impractical for resource-constrained devices like s...

Full description

Saved in:
Bibliographic Details
Main Authors: Nadeem Jabbar, Sohail Masood Bhatti, Muhammad Rashid, Arfan Jaffar, Sheeraz Akram
Format: Article
Language:English
Published: Public Library of Science (PLoS) 2025-01-01
Series:PLoS ONE
Online Access:https://doi.org/10.1371/journal.pone.0326565
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849721148143566848
author Nadeem Jabbar
Sohail Masood Bhatti
Muhammad Rashid
Arfan Jaffar
Sheeraz Akram
author_facet Nadeem Jabbar
Sohail Masood Bhatti
Muhammad Rashid
Arfan Jaffar
Sheeraz Akram
author_sort Nadeem Jabbar
collection DOAJ
description Deepfakes, synthetic media created using artificial intelligence, threaten the authenticity of digital content. Traditional detection methods, such as Convolutional Neural Networks (CNNs), require substantial computational resources, rendering them impractical for resource-constrained devices like smartphones and IoT systems. This study evaluates a single-layer Kolmogorov-Arnold Network (KAN) with 200 nodes for efficient deepfake classification. Experimental results show that KAN achieves 95.01% accuracy on the FaceForensics++ dataset and 88.32% on the Celeb-DF dataset, while requiring only 52.4 MB of memory, 13.11 million parameters, and 26.21 million FLOPs, significantly less than state-of-the-art CNNs. These verified metrics highlight KAN's potential for real-time deepfake detection on edge devices. Untested capabilities, such as robustness against adversarial attacks, are proposed for future research. This work aligns with the United Nations Sustainable Development Goals, specifically SDG 9: Industry, Innovation, and Infrastructure and SDG 16: Peace, Justice, and Strong Institutions.
format Article
id doaj-art-9cd2883fc362469e9294ddce7492b9fb
institution DOAJ
issn 1932-6203
language English
publishDate 2025-01-01
publisher Public Library of Science (PLoS)
record_format Article
series PLoS ONE
spelling doaj-art-9cd2883fc362469e9294ddce7492b9fb2025-08-20T03:11:46ZengPublic Library of Science (PLoS)PLoS ONE1932-62032025-01-01207e032656510.1371/journal.pone.0326565Single-layer KAN for deepfake classification: Balancing efficiency and performance in resource constrained environments.Nadeem JabbarSohail Masood BhattiMuhammad RashidArfan JaffarSheeraz AkramDeepfakes, synthetic media created using artificial intelligence, threaten the authenticity of digital content. Traditional detection methods, such as Convolutional Neural Networks (CNNs), require substantial computational resources, rendering them impractical for resource-constrained devices like smartphones and IoT systems. This study evaluates a single-layer Kolmogorov-Arnold Network (KAN) with 200 nodes for efficient deepfake classification. Experimental results show that KAN achieves 95.01% accuracy on the FaceForensics++ dataset and 88.32% on the Celeb-DF dataset, while requiring only 52.4 MB of memory, 13.11 million parameters, and 26.21 million FLOPs, significantly less than state-of-the-art CNNs. These verified metrics highlight KAN's potential for real-time deepfake detection on edge devices. Untested capabilities, such as robustness against adversarial attacks, are proposed for future research. This work aligns with the United Nations Sustainable Development Goals, specifically SDG 9: Industry, Innovation, and Infrastructure and SDG 16: Peace, Justice, and Strong Institutions.https://doi.org/10.1371/journal.pone.0326565
spellingShingle Nadeem Jabbar
Sohail Masood Bhatti
Muhammad Rashid
Arfan Jaffar
Sheeraz Akram
Single-layer KAN for deepfake classification: Balancing efficiency and performance in resource constrained environments.
PLoS ONE
title Single-layer KAN for deepfake classification: Balancing efficiency and performance in resource constrained environments.
title_full Single-layer KAN for deepfake classification: Balancing efficiency and performance in resource constrained environments.
title_fullStr Single-layer KAN for deepfake classification: Balancing efficiency and performance in resource constrained environments.
title_full_unstemmed Single-layer KAN for deepfake classification: Balancing efficiency and performance in resource constrained environments.
title_short Single-layer KAN for deepfake classification: Balancing efficiency and performance in resource constrained environments.
title_sort single layer kan for deepfake classification balancing efficiency and performance in resource constrained environments
url https://doi.org/10.1371/journal.pone.0326565
work_keys_str_mv AT nadeemjabbar singlelayerkanfordeepfakeclassificationbalancingefficiencyandperformanceinresourceconstrainedenvironments
AT sohailmasoodbhatti singlelayerkanfordeepfakeclassificationbalancingefficiencyandperformanceinresourceconstrainedenvironments
AT muhammadrashid singlelayerkanfordeepfakeclassificationbalancingefficiencyandperformanceinresourceconstrainedenvironments
AT arfanjaffar singlelayerkanfordeepfakeclassificationbalancingefficiencyandperformanceinresourceconstrainedenvironments
AT sheerazakram singlelayerkanfordeepfakeclassificationbalancingefficiencyandperformanceinresourceconstrainedenvironments