Classification vs regression in overparameterized regimes: Does the loss function matter?
Vidya MuthukumarAdhyyan NarangVignesh SubramanianMikhail BelkinDaniel HsuAnant SahaiPublished in: J. Mach. Learn. Res. (2021)
Keyphrases
- loss function
- support vector
- hinge loss
- reproducing kernel hilbert space
- gradient boosting
- classification accuracy
- pairwise
- learning to rank
- machine learning
- pattern classification
- decision trees
- model selection
- feature space
- support vector machine
- logistic regression
- regression problems
- boosting framework
- benchmark datasets
- class labels
- supervised learning
- feature vectors
- linear regression
- text classification
- boosting algorithms
- base learners
- risk minimization
- learning algorithm
- feature selection
- pairwise constraints
- stochastic gradient descent
- bayes rule
- support vector machine svm
- feature set