Goto

Collaborating Authors

 version space




TowardsaUnified Information-Theoretic FrameworkforGeneralization

Neural Information Processing Systems

Let D be an unknown distribution on a spaceZ, and let H be a set of classifiers. Consider a (randomized) learning algorithmA = (An)n 1 that selects an elementˆh in H, based onn i.i.d.






6a26c75d6a576c94654bfc4dda548c72-Paper.pdf

Neural Information Processing Systems

Forlinear regression, we give a polynomial-time algorithm based on Celis-Dennis-Tapia optimization algorithms. For binary classification, we show how to efficiently implement itusing aproper agnostic learner (i.e., anEmpirical Risk Minimizer) for the class of interest.