Peter Grünwald OER/OCW Courses

Bounds on individual risk for log-loss by Peter Grünwald @VideoLectures

In sequential prediction with log-loss as well as density estimation with risk measured by KL divergence, one is often interested in the expected instantaneous loss, or, equivalently... Watch Video

MDL Tutorial by Peter Grünwald @VideoLectures.net oer ocw free Open

We give a self-contained tutorial on the Minimum Description Length (MDL) approach to modeling, learning and prediction. We focus on the recent (post 1995) formulations of MDL, which... Watch Video

Safe Learning: bridging the gap between by Peter Grünwald @VideoLectures

We extend Bayesian MAP and Minimum Description Length (MDL) learning by testing whether the data can be substantially more compressed by a mixture of the MDL/MAP distribution with... Watch Video

Suboptimality of MDL and Bayes in by Peter Grünwald @VideoLectures

We show that forms of Bayesian and MDL learning that are often applied to classification problems can be *statistically inconsistent*. We present a large family of classifiers and... Watch Video

The Catch-Up Phenomenon in Bayesian by Peter Grünwald @VideoLectures

Standard Bayesian model selection/averaging sometimes learn too slowly: there exist other learning methods that lead to better predictions based on less data. We give a novel analysis... Watch Video

Universal Coding/Prediction and Statistical by Peter @VideoLectures

Part of this talk is based on results of A. Barron (1986) and recent joint work with J. Langford (2004). We introduce the information-theoretic concepts of universal coding and... Watch Video

Universal Modeling: Introduction to modern by Peter @VideoLectures

We give a tutorial introduction to the *modern* Minimum Description Length (MDL) Principle, taking into account the many refinements and developments that have taken place in the 1... Watch Video

We need a BIT more GUTS Grand Unified by Peter Grünwald @VideoLectures

A remarkable variety of problems in machine learning and statistics can be recast as data compression under constraints: (1) sequential prediction with arbitrary loss functions can... Watch Video

Find OpenCourseWare Online Exams!
Attribution: The Open Education Consortium http://www.oeconsortium.org