Are Straight-Through gradients and Soft-Thresholding all you need for Sparse Training?
Antoine VanderschuerenChristophe De VleeschouwerPublished in: CoRR (2022)
Keyphrases
- soft thresholding
- image denoising
- total variation
- wavelet domain
- denoising
- high dimensional
- avoid overfitting
- sparse data
- test set
- noise reduction
- training set
- training algorithm
- online learning
- line segments
- e learning
- straight line
- training phase
- sparse representation
- multiresolution
- neural network
- image quality
- support vector machine
- training data
- compressed sensing
- image segmentation
- machine learning