Correlation Analysis of Adversarial Attack in Time Series Classification
This work addresses the vulnerability of time series classification models to adversarial attacks, offering insights for designing more resilient neural networks, though it appears incremental by building on existing frequency domain analysis.
This study investigated how time series classification models process local versus global information under adversarial attacks, finding that regularization techniques using FFT methods enhance attack effectiveness while defense strategies like noise introduction and Gaussian filtering lower Attack Success Rate, with models prioritizing global information showing greater resistance.
This study investigates the vulnerability of time series classification models to adversarial attacks, with a focus on how these models process local versus global information under such conditions. By leveraging the Normalized Auto Correlation Function (NACF), an exploration into the inclination of neural networks is conducted. It is demonstrated that regularization techniques, particularly those employing Fast Fourier Transform (FFT) methods and targeting frequency components of perturbations, markedly enhance the effectiveness of attacks. Meanwhile, the defense strategies, like noise introduction and Gaussian filtering, are shown to significantly lower the Attack Success Rate (ASR), with approaches based on noise introducing notably effective in countering high-frequency distortions. Furthermore, models designed to prioritize global information are revealed to possess greater resistance to adversarial manipulations. These results underline the importance of designing attack and defense mechanisms, informed by frequency domain analysis, as a means to considerably reinforce the resilience of neural network models against adversarial threats.