Regularization-wise double descent: Why it occurs and how to eliminate it.
Fatih Furkan YilmazReinhard HeckelPublished in: CoRR (2022)
Keyphrases
- pairwise
- regularization method
- sparsity regularization
- regularization parameter
- prior information
- regularization methods
- blind deconvolution
- parameter selection
- data dependent
- regularization term
- neural network
- artificial intelligence
- probability distribution
- edge preserving
- special case
- active learning
- prior knowledge
- reproducing kernel hilbert space
- expert systems
- bayesian networks
- genetic algorithm