Regularization-wise double descent: Why it occurs and how to eliminate it.
Fatih Furkan YilmazReinhard HeckelPublished in: ISIT (2022)
Keyphrases
- regularization method
- prior information
- regularization parameter
- empirical risk minimization
- expert systems
- pairwise
- data dependent
- smoothing parameter
- information retrieval
- regularization framework
- inverse problems
- reproducing kernel hilbert space
- parameter selection
- regularization term
- model selection
- maximum likelihood
- objective function
- multiscale
- computer vision
- learning algorithm