Detecting chaos in adversarial examples

The puzzling phenomenon of adversarial examples continues to attract significant research within the machine learning community. The confirmation that adversarial examples can arise in natural real-life circumstances has but increased the interest. While several methods have been proposed for both g...

Descripción completa

Detalles Bibliográficos
Autores: Déniz Suárez, Óscar, Pedraza Dorado, Aníbal, Bueno García, María Gloria
Tipo de recurso: artículo
Fecha de publicación:2022
País:España
Institución:Universidad de Castilla-La Mancha
Repositorio:RUIdeRA. Repositorio Institucional de la UCLM
OAI Identifier:oai:ruidera.uclm.es:10578/45779
Acceso en línea:https://doi.org/10.1016/j.chaos.2022.112577
https://www.sciencedirect.com/science/article/pii/S0960077922007676
https://hdl.handle.net/10578/45779
Access Level:acceso abierto
Palabra clave:Adversarial examples
Chaos theory
Lyapunov stability
Neural networksDeep learning
Trustworthy machine learning
Descripción
Sumario:The puzzling phenomenon of adversarial examples continues to attract significant research within the machine learning community. The confirmation that adversarial examples can arise in natural real-life circumstances has but increased the interest. While several methods have been proposed for both generating adversarial examples and defending against them, in this work we focus on a previous serendipitous discovery indicating that they can be considered as chaotic signals. More specifically, it has been recently shown that measures of chaoticity in the input signal can be used to detect adversarial examples efficiently. In this work, we extend that approach in two aspects, leading to significant improvements in detection accuracy as demonstrated by results obtained in experiments with four datasets and using seven different attack methods.