A spectral analysis framework for studying aliasing in strided CNNs. Defines the Alias Violation Ratio (AVR) and Shift Instability Score (SIS) to quantify Nyquist violations and their relationship to prediction instability.
computer-vision deep-learning signal-processing pytorch resnet convolutional-neural-networks spectral-analysis frequency-analysis fourier-transform anti-aliasing aliasing nyquist-theorem shift-invariance blurpool
-
Updated
May 11, 2026 - Python