On the Convergence Speed of MDL Predictions for Bernoulli Sequences
We consider the Minimum Description Length principle for online sequence prediction. If the underlying model class is discrete, then the total expected square loss is a particularly interesting performance measure: (a) this quantity is bounded, implying convergence with probability one, and (b) it additionally specifies a rate of convergence. Generally, for MDL only exponential loss bounds hold, as opposed to the linear bounds for a Bayes mixture. We show that this is even the case if the model class contains only Bernoulli distributions. We derive a new upper bound on the prediction error for countable Bernoulli classes. This implies a small bound (comparable to the one for Bayes mixtures) for certain important model classes. The results apply to many Machine Learning tasks including classification and hypothesis testing. We provide arguments that our theorems generalize to countable classes of i.i.d. models.
KeywordsConvergence Speed True Parameter Kolmogorov Complexity Cumulative Loss Universal Turing Machine
Unable to display preview. Download preview PDF.
- 12.Hutter, M.: Sequence prediction based on monotone complexity. In: Proc. 16th Annual Conference on Learning Theory (COLT 2003). LNCS (LNAI), pp. 506–521. Springer, Berlin (2003)Google Scholar
- 14.Hutter, M.: Sequential predictions based on algorithmic complexity. Technical report (2004) IDSIA-16-04Google Scholar