Work

Quantum adversarial learning for kernel methods

Year: 2025

Type: article

Abstract: We show that hybrid quantum classifiers based on quantum kernel methods and support vector machines are vulnerable against adversarial attacks, namely small engineered perturbations of the input data can deceive the classifier into predicting the wrong result. Nonetheless, we also show that simple defense strategies based on data augmentation with a few crafted perturbations can make the classifier robust against new attacks. Our results find applications in security-critical learning problems and in mitigating the effect of some forms of quantum noise, since the attacker can also be understood as part of the surrounding environment.

Source: Quantum Machine Intelligence

DOI: 10.1007/s42484-025-00238-8

Authors: Giuseppe Montalbano, Leonardo Banchi

Institutions: Ca' Foscari University of Venice, Istituto Nazionale di Fisica Nucleare, Sezione di Firenze, University of Florence

Language: English

Topic: Adversarial Robustness in Machine Learning

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: hybrid

Estimated APC paid: $3,390

Funders: U.S. Department of Energy, Università degli Studi di Firenze, Ministero dell'Università e della Ricerca, Office of Science