Material Detail

Safe Learning: bridging the gap between Bayes, MDL and statistical learning theory via empirical convexity

Safe Learning: bridging the gap between Bayes, MDL and statistical learning theory via empirical convexity

This video was recorded at 24th Annual Conference on Learning Theory (COLT), Budapest 2011. 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 another element of the model, and adjusting the learning rate if this is the case. While standard Bayes and MDL can fail to converge if the model is wrong, the resulting "safe" estimator continues to achieve good rates with wrong models. Moreover, when applied to classification and regression models as considered in statistical learning theory, the approach achieves optimal rates under, e.g., Tsybakov's conditions, and reveals new situations in which we can penalize by (-log PRIOR)/n rather than √(-log PRIOR)/n.
Rate

Quality

  • User Rating
  • Comments
  • Learning Exercises
  • Bookmark Collections
  • Course ePortfolios
  • Accessibility Info

More about this material

Browse...

Disciplines with similar materials as Safe Learning: bridging the gap between Bayes, MDL and statistical learning theory via empirical convexity

Comments

Log in to participate in the discussions or sign up if you are not already a MERLOT member.
hidden