A Systematic Review on Model Watermarking for Neural Networks

Machine learning (ML) models are applied in an increasing variety of domains. The availability of large amounts of data and computational resources encourages the development of ever more complex and valuable models. These models are considered the intellectual property of the legitimate parties who...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autor principal:	Franziska Boenisch
Formato:	article
Lenguaje:	EN
Publicado:	Frontiers Media S.A. 2021
Materias:	neural networks intellectual property protection watermarking machine learning model stealing Information technology T58.5-58.64
Acceso en línea:	https://doaj.org/article/429416b4de194c82b7b315fde8dd04b0
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:429416b4de194c82b7b315fde8dd04b0
record_format	dspace
spelling	oai:doaj.org-article:429416b4de194c82b7b315fde8dd04b02021-12-01T13:40:01ZA Systematic Review on Model Watermarking for Neural Networks2624-909X10.3389/fdata.2021.729663https://doaj.org/article/429416b4de194c82b7b315fde8dd04b02021-11-01T00:00:00Zhttps://www.frontiersin.org/articles/10.3389/fdata.2021.729663/fullhttps://doaj.org/toc/2624-909XMachine learning (ML) models are applied in an increasing variety of domains. The availability of large amounts of data and computational resources encourages the development of ever more complex and valuable models. These models are considered the intellectual property of the legitimate parties who have trained them, which makes their protection against stealing, illegitimate redistribution, and unauthorized application an urgent need. Digital watermarking presents a strong mechanism for marking model ownership and, thereby, offers protection against those threats. This work presents a taxonomy identifying and analyzing different classes of watermarking schemes for ML models. It introduces a unified threat model to allow structured reasoning on and comparison of the effectiveness of watermarking methods in different scenarios. Furthermore, it systematizes desired security requirements and attacks against ML model watermarking. Based on that framework, representative literature from the field is surveyed to illustrate the taxonomy. Finally, shortcomings and general limitations of existing approaches are discussed, and an outlook on future research directions is given.Franziska BoenischFrontiers Media S.A.articleneural networksintellectual property protectionwatermarkingmachine learningmodel stealingInformation technologyT58.5-58.64ENFrontiers in Big Data, Vol 4 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	neural networks intellectual property protection watermarking machine learning model stealing Information technology T58.5-58.64
spellingShingle	neural networks intellectual property protection watermarking machine learning model stealing Information technology T58.5-58.64 Franziska Boenisch A Systematic Review on Model Watermarking for Neural Networks
description	Machine learning (ML) models are applied in an increasing variety of domains. The availability of large amounts of data and computational resources encourages the development of ever more complex and valuable models. These models are considered the intellectual property of the legitimate parties who have trained them, which makes their protection against stealing, illegitimate redistribution, and unauthorized application an urgent need. Digital watermarking presents a strong mechanism for marking model ownership and, thereby, offers protection against those threats. This work presents a taxonomy identifying and analyzing different classes of watermarking schemes for ML models. It introduces a unified threat model to allow structured reasoning on and comparison of the effectiveness of watermarking methods in different scenarios. Furthermore, it systematizes desired security requirements and attacks against ML model watermarking. Based on that framework, representative literature from the field is surveyed to illustrate the taxonomy. Finally, shortcomings and general limitations of existing approaches are discussed, and an outlook on future research directions is given.
format	article
author	Franziska Boenisch
author_facet	Franziska Boenisch
author_sort	Franziska Boenisch
title	A Systematic Review on Model Watermarking for Neural Networks
title_short	A Systematic Review on Model Watermarking for Neural Networks
title_full	A Systematic Review on Model Watermarking for Neural Networks
title_fullStr	A Systematic Review on Model Watermarking for Neural Networks
title_full_unstemmed	A Systematic Review on Model Watermarking for Neural Networks
title_sort	systematic review on model watermarking for neural networks
publisher	Frontiers Media S.A.
publishDate	2021
url	https://doaj.org/article/429416b4de194c82b7b315fde8dd04b0
work_keys_str_mv	AT franziskaboenisch asystematicreviewonmodelwatermarkingforneuralnetworks AT franziskaboenisch systematicreviewonmodelwatermarkingforneuralnetworks
_version_	1718405113107709952

A Systematic Review on Model Watermarking for Neural Networks

Ejemplares similares