Adversarial Vulnerabilities of Neural Biomarker Identification Systems
Abstract
There is growing interest in the proposed use of EEG signals as biometric credentials, but thus far there has been little research on the reliability and security of such biometrics. Prior adversarial tests have focused on deep-learning classifiers and assumed attackers have full access to the classifier model. This has left unexamined other, more popular categories of neural signature methods as well as the more realistic case of an adversary having only black-box access to a classifier. In this paper we develop a collection of adaptive attack algorithms which learn to fool an authentication system via targeted alterations of stolen EEG recordings, without requiring any knowledge of the authentication system itself. Tested on 6 public datasets spanning three recording conditions (reacting to visual stimuli, imagining hand movements, and resting), it reveals that different signaturing approaches vary significantly in their degrees of vulnerability to adversarial attacks. We show that vulnerability to spoofing attack is greatly impacted by the recording conditions, with significant variation depending on task at time of recording. Finally, we provide recommendations for improving neural signature biometrics based on the results of our adversarial testing.
Related
- Identification of Defined Molecular Subgroups on the Basis of Immunohistochemical Analyses and Potential Therapeutic Vulnerabilities of Pulmonary Carcinoids
- Neural prediction decorrelation reveals that adversarial robustness substantially improves DNN prediction accuracy across the entire human auditory cortex
- Explainable computational framework for subtype classification and biomarker identification in head and neck squamous cell carcinoma using machine learning
- Identification of Microalgae, Evaluation of Water Purification Efficiency, and Optimization of Large-Scale Production Conditions Based on Recirculating Aquaculture Systems
- Hypothesis-Driven Identification of Neural Algorithms With Dynamical Structure-Preserving Manifolds.
BibTeX
@misc{tapal2026adversarial,
title = {{Adversarial Vulnerabilities of Neural Biomarker Identification Systems}},
author = {Polina Tapal and Bryce-Allen Bagley},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.01856},
eprint = {2609.01856},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.01856},
}
RIS
TY - UNPB
TI - Adversarial Vulnerabilities of Neural Biomarker Identification Systems
AU - Polina Tapal
AU - Bryce-Allen Bagley
PY - 2026
DA - 2026/09/01
PB - arXiv
DO - 10.48550/arxiv.2609.01856
UR - https://arxiv.org/abs/2609.01856
AB - There is growing interest in the proposed use of EEG signals as biometric credentials, but thus far there has been little research on the reliability and security of such biometrics. Prior adversarial tests have focused on deep-learning classifiers and assumed attackers have full access to the classifier model. This has left unexamined other, more popular categories of neural signature methods as well as the more realistic case of an adversary having only black-box access to a classifier. In this paper we develop a collection of adaptive attack algorithms which learn to fool an authentication system via targeted alterations of stolen EEG recordings, without requiring any knowledge of the authentication system itself. Tested on 6 public datasets spanning three recording conditions (reacting to visual stimuli, imagining hand movements, and resting), it reveals that different signaturing approaches vary significantly in their degrees of vulnerability to adversarial attacks. We show that vulnerability to spoofing attack is greatly impacted by the recording conditions, with significant variation depending on task at time of recording. Finally, we provide recommendations for improving neural signature biometrics based on the results of our adversarial testing.
ER -