Week 3 Flashcards
What is the definition of emperical risk?
What is the definition of a unbiased risk estimator?
What is a naive approach to emperical risk minimization?
What is the definition of a model and a Emperical Risk Minimizer?
What can be done to avoid the naive approach?
Reframe a linear regression as an ERM.
What is the definition of a partition?
What is the definition of a step-function over a partition?
What do you need to proof when finding the ERM of a step-function for regression.
Proof the ERM of a step-function of a regression.
What do needs to proved when proving the ERM of a step-function for classification?
Proof the ERM of a step-function of a classification.
What is the definition of a learning rule?
What is the definition of a ERM learning rule?
What should a good learning rule have?
How are classification learning rules and regression learning rules linked?
Why is a 0-1 cost function a bad idea?
What is the definition of a pseudo-classifier?
What is the definition of a plug-in classifier
What is the definition of a Empirical Convexified Risk Minimizer?
What is the definition of Plug-in classification rule associated to ECRM.
What is the idea of convex surrogate loss?
Proof that a good regression is also a good classifier.
