PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Mixability is Bayes Risk Curvature Relative to Log Loss
Tim Erven, van, Mark Reid and Bob Williamson
Journal of Machine Learning Research 2011.

Abstract

Mixability of a loss governs the best possible performance when aggregating expert predictions with respect to that loss. The determination of the mixability constant for binary losses is straightforward but opaque. In the binary case we make this transparent and simpler by characterising mixability in terms of the second derivative of the Bayes risk of proper losses. We then extend this result to multiclass proper losses where there are few existing results. We show that mixability is governed by the maximum eigenvalue of the Hessian of the Bayes risk, relative to the Hessian of the Bayes risk for log loss. We conclude by comparing our result to other work that bounds prediction performance in terms of the geometry of the Bayes risk. Although all calculations are for proper losses, we also show how to carry the results across to improper losses.

EPrint Type:Article
Additional Information:Accepted pending a minor revision
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Computational, Information-Theoretic Learning with Statistics
ID Code:8929
Deposited By:Tim Erven, van
Deposited On:21 February 2012