Perturbing the Phase: Analyzing Adversarial Robustness of Complex-Valued Neural Networks
This work addresses the safety and robustness of CVNNs for practical applications, but it is incremental as it extends adversarial attack analysis to complex-valued domains.
The paper tackled the problem of adversarial robustness in complex-valued neural networks (CVNNs) by designing Phase Attacks targeting phase information and adapting existing attacks to complex values, showing that CVNNs can be more robust in some scenarios but are highly susceptible to phase changes, with Phase Attacks reducing model performance more than regular attacks.
Complex-valued neural networks (CVNNs) are rising in popularity for all kinds of applications. To safely use CVNNs in practice, analyzing their robustness against outliers is crucial. One well known technique to understand the behavior of deep neural networks is to investigate their behavior under adversarial attacks, which can be seen as worst case minimal perturbations. We design Phase Attacks, a kind of attack specifically targeting the phase information of complex-valued inputs. Additionally, we derive complex-valued versions of commonly used adversarial attacks. We show that in some scenarios CVNNs are more robust than RVNNs and that both are very susceptible to phase changes with the Phase Attacks decreasing the model performance more, than equally strong regular attacks, which can attack both phase and magnitude.