Enhancing Security of Proof-of-Learning Against Spoofing Attacks Using Feature-Based Model Watermarking

The rapid advancement of machine learning (ML) technologies necessitates robust security frameworks to protect the integrity of ML model training processes. Proof-of-Learning (PoL) is a critical method for verifying the computational effort in training ML models, while model watermarking is a strate...

Full description

Saved in:
Bibliographic Details
Main Authors: Ozgur Ural, Kenji Yoshigoe
Format: Article
Language:English
Published: IEEE 2024-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10741282/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850130254843084800
author Ozgur Ural
Kenji Yoshigoe
author_facet Ozgur Ural
Kenji Yoshigoe
author_sort Ozgur Ural
collection DOAJ
description The rapid advancement of machine learning (ML) technologies necessitates robust security frameworks to protect the integrity of ML model training processes. Proof-of-Learning (PoL) is a critical method for verifying the computational effort in training ML models, while model watermarking is a strategy for asserting model ownership. This research integrates PoL with feature-based model watermarking, embedding the watermark directly into the model’s features or parameters. This integration mitigates security risks associated with external key management and reduces computational overhead by eliminating the need for complex verification procedures. Our proposed dual-layered verification architecture embeds unique watermarks during the training phase. It records them alongside PoL proofs, enhancing security against sophisticated spoofing attacks where adversaries attempt to mimic a model’s computational trajectory and watermark. This approach addresses critical challenges, including maintaining watermark robustness and balancing security with model performance. Through a comprehensive analysis, we identify vulnerabilities in existing PoL systems and demonstrate how feature-based watermarking can enhance security. We present a secure PoL mechanism, supported by empirical validation, that significantly improves resilience to spoofing attacks. This advancement represents a crucial step towards securing ML models, paving the way for future research to protect diverse ML applications from various threats.
format Article
id doaj-art-e31b03075d234cd98812c6566153f07f
institution OA Journals
issn 2169-3536
language English
publishDate 2024-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-e31b03075d234cd98812c6566153f07f2025-08-20T02:32:44ZengIEEEIEEE Access2169-35362024-01-011216956716959110.1109/ACCESS.2024.348977610741282Enhancing Security of Proof-of-Learning Against Spoofing Attacks Using Feature-Based Model WatermarkingOzgur Ural0https://orcid.org/0000-0003-1329-4303Kenji Yoshigoe1https://orcid.org/0000-0001-6040-4742Department of Electrical Engineering and Computer Science, Embry-Riddle Aeronautical University, Daytona Beach, FL, USADepartment of Electrical Engineering and Computer Science, Embry-Riddle Aeronautical University, Daytona Beach, FL, USAThe rapid advancement of machine learning (ML) technologies necessitates robust security frameworks to protect the integrity of ML model training processes. Proof-of-Learning (PoL) is a critical method for verifying the computational effort in training ML models, while model watermarking is a strategy for asserting model ownership. This research integrates PoL with feature-based model watermarking, embedding the watermark directly into the model’s features or parameters. This integration mitigates security risks associated with external key management and reduces computational overhead by eliminating the need for complex verification procedures. Our proposed dual-layered verification architecture embeds unique watermarks during the training phase. It records them alongside PoL proofs, enhancing security against sophisticated spoofing attacks where adversaries attempt to mimic a model’s computational trajectory and watermark. This approach addresses critical challenges, including maintaining watermark robustness and balancing security with model performance. Through a comprehensive analysis, we identify vulnerabilities in existing PoL systems and demonstrate how feature-based watermarking can enhance security. We present a secure PoL mechanism, supported by empirical validation, that significantly improves resilience to spoofing attacks. This advancement represents a crucial step towards securing ML models, paving the way for future research to protect diverse ML applications from various threats.https://ieeexplore.ieee.org/document/10741282/Proof-of-learningmodel watermarkingmachine learning securityspoofing attack countermeasuresdual-layered verificationmodel authenticity
spellingShingle Ozgur Ural
Kenji Yoshigoe
Enhancing Security of Proof-of-Learning Against Spoofing Attacks Using Feature-Based Model Watermarking
IEEE Access
Proof-of-learning
model watermarking
machine learning security
spoofing attack countermeasures
dual-layered verification
model authenticity
title Enhancing Security of Proof-of-Learning Against Spoofing Attacks Using Feature-Based Model Watermarking
title_full Enhancing Security of Proof-of-Learning Against Spoofing Attacks Using Feature-Based Model Watermarking
title_fullStr Enhancing Security of Proof-of-Learning Against Spoofing Attacks Using Feature-Based Model Watermarking
title_full_unstemmed Enhancing Security of Proof-of-Learning Against Spoofing Attacks Using Feature-Based Model Watermarking
title_short Enhancing Security of Proof-of-Learning Against Spoofing Attacks Using Feature-Based Model Watermarking
title_sort enhancing security of proof of learning against spoofing attacks using feature based model watermarking
topic Proof-of-learning
model watermarking
machine learning security
spoofing attack countermeasures
dual-layered verification
model authenticity
url https://ieeexplore.ieee.org/document/10741282/
work_keys_str_mv AT ozgurural enhancingsecurityofproofoflearningagainstspoofingattacksusingfeaturebasedmodelwatermarking
AT kenjiyoshigoe enhancingsecurityofproofoflearningagainstspoofingattacksusingfeaturebasedmodelwatermarking