Rademacher complexity is said to be the most powerfull measure of model capacity in statistical learning theory. But in statistical learning class we are just focusing on Rademacher complexity of linear predictors. I am wondering if we could apply Rademacher complexity to any random hypothesis class, like decision tree, k-NN and SVM.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.