A Backdoor Approach With Inverted Labels Using Dirty Label-Flipping Attacks

Audio-based machine learning systems frequently use public or third-party data, which might be inaccurate. This exposes deep neural network (DNN) models trained on such data to potential data poisoning attacks. In this type of assault, attackers can train the DNN model using poisoned data, potential...

Full description

Saved in:
Bibliographic Details
Main Author: Orson Mengara
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10483076/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849317686323970048
author Orson Mengara
author_facet Orson Mengara
author_sort Orson Mengara
collection DOAJ
description Audio-based machine learning systems frequently use public or third-party data, which might be inaccurate. This exposes deep neural network (DNN) models trained on such data to potential data poisoning attacks. In this type of assault, attackers can train the DNN model using poisoned data, potentially degrading its performance. Another type of data poisoning attack that is extremely relevant to our investigation is label flipping, in which the attacker manipulates the labels for a subset of data. It has been demonstrated that these assaults may drastically reduce system performance, even for attackers with minimal abilities. In this study, we propose a backdoor attack named “DirtyFlipping”, which uses dirty label techniques, ‘label-on-label‘, to input triggers (clapping) in the selected data patterns associated with the target class, thereby enabling a stealthy backdoor.
format Article
id doaj-art-c370d42a212945c68356c1f9e586c195
institution Kabale University
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-c370d42a212945c68356c1f9e586c1952025-08-20T03:51:08ZengIEEEIEEE Access2169-35362025-01-011312422512423310.1109/ACCESS.2024.338283910483076A Backdoor Approach With Inverted Labels Using Dirty Label-Flipping AttacksOrson Mengara0https://orcid.org/0009-0009-4022-3499INRS-EMT, University of Quebec, Montreal, QC, CanadaAudio-based machine learning systems frequently use public or third-party data, which might be inaccurate. This exposes deep neural network (DNN) models trained on such data to potential data poisoning attacks. In this type of assault, attackers can train the DNN model using poisoned data, potentially degrading its performance. Another type of data poisoning attack that is extremely relevant to our investigation is label flipping, in which the attacker manipulates the labels for a subset of data. It has been demonstrated that these assaults may drastically reduce system performance, even for attackers with minimal abilities. In this study, we propose a backdoor attack named “DirtyFlipping”, which uses dirty label techniques, ‘label-on-label‘, to input triggers (clapping) in the selected data patterns associated with the target class, thereby enabling a stealthy backdoor.https://ieeexplore.ieee.org/document/10483076/Poisoning attacksbackdoor attacksadversarial machine learning
spellingShingle Orson Mengara
A Backdoor Approach With Inverted Labels Using Dirty Label-Flipping Attacks
IEEE Access
Poisoning attacks
backdoor attacks
adversarial machine learning
title A Backdoor Approach With Inverted Labels Using Dirty Label-Flipping Attacks
title_full A Backdoor Approach With Inverted Labels Using Dirty Label-Flipping Attacks
title_fullStr A Backdoor Approach With Inverted Labels Using Dirty Label-Flipping Attacks
title_full_unstemmed A Backdoor Approach With Inverted Labels Using Dirty Label-Flipping Attacks
title_short A Backdoor Approach With Inverted Labels Using Dirty Label-Flipping Attacks
title_sort backdoor approach with inverted labels using dirty label flipping attacks
topic Poisoning attacks
backdoor attacks
adversarial machine learning
url https://ieeexplore.ieee.org/document/10483076/
work_keys_str_mv AT orsonmengara abackdoorapproachwithinvertedlabelsusingdirtylabelflippingattacks
AT orsonmengara backdoorapproachwithinvertedlabelsusingdirtylabelflippingattacks