The Optimality of Polynomial Regression for Agnostic Learning under Gaussian Marginals.
Ilias DiakonikolasDaniel M. KaneThanasis PittasNikos ZarifisPublished in: CoRR (2021)
Keyphrases
- polynomial regression
- agnostic learning
- uniform distribution
- locally weighted
- noise tolerant
- membership queries
- pac learning
- mdl principle
- probability distribution
- boosting algorithms
- concept class
- optimal solution
- regression model
- maximum likelihood
- linear regression
- target function
- active learning
- gaussian distribution
- binary classification problems
- learning algorithm
- multi class
- feature vectors
- lower bound
- support vector
- decision trees