Dissecting the impact of different loss functions with gradient surgery.
Hong XuanRobert PlessPublished in: CoRR (2022)
Keyphrases
- loss function
- pairwise
- support vector
- risk minimization
- logistic regression
- learning to rank
- squared error
- loss minimization
- empirical risk
- reproducing kernel hilbert space
- convex loss functions
- learning algorithm
- hinge loss
- boosting framework
- learning models
- update rules
- machine learning
- learning theory
- information retrieval