On the Structural Sensitivity of Deep Convolutional Networks to the Directions of Fourier Basis Functions.
Yusuke TsuzukuIssei SatoPublished in: CVPR (2019)
Keyphrases
- basis functions
- series expansion
- linear combination
- hilbert transform
- deep learning
- b spline
- state space
- radial basis function
- nonlinear functions
- spherical harmonics
- fourier transform
- matching pursuit
- signal reconstruction
- radon transform
- zernike moments
- linear approximation
- unsupervised feature learning
- factored mdps
- set of basis functions
- neural network
- sparse coding
- multiscale
- deep belief networks
- image reconstruction
- lookup tables
- frequency domain
- multiresolution
- learning algorithm