Investigating imperceptibility of adversarial attacks on tabular data: An empirical analysis

Zhipeng He, Chun Ouyang, Laith Alzubaidi, Alistair Barros, Catarina Moreira

Abstract: Highlights•Seven properties to characterise imperceptible adversarial attacks on tabular data.•Evaluation of adversarial attacks’ imperceptibility using the proposed metrics.•Evaluation reveals a trade-off between attacks’ imperceptibility and effectiveness.•Addressing limitations in attack algorithms to improve imperceptibility.