How regularization affects the geometry of loss functions.
Nathaniel BottmanY. CooperAntonio LerarioPublished in: CoRR (2023)
Keyphrases
- loss function
- reproducing kernel hilbert space
- convex loss functions
- risk minimization
- hinge loss
- loss minimization
- gradient boosting
- pairwise
- square loss
- support vector
- empirical risk
- bregman divergences
- logistic regression
- learning to rank
- squared error
- uniform convergence
- parameter selection
- gaussian kernels
- precision and recall
- boosting algorithms
- update rules
- training samples
- aggregating algorithm
- semi supervised