Context Tree Switching

TL;DR

The paper introduces Context Tree Switching, a modification of Context Tree Weighting that improves prediction accuracy for binary stationary sources without increasing computational complexity.

Contribution

It presents a novel model mixing technique that extends Context Tree Weighting's class of models while maintaining its efficiency and theoretical guarantees.

Findings

Empirically outperforms Context Tree Weighting on the Calgary Corpus.

Preserves theoretical properties for stationary n-Markov sources.

Maintains asymptotic time and space complexity.

Abstract

This paper describes the Context Tree Switching technique, a modification of Context Tree Weighting for the prediction of binary, stationary, n-Markov sources. By modifying Context Tree Weighting's recursive weighting scheme, it is possible to mix over a strictly larger class of models without increasing the asymptotic time or space complexity of the original algorithm. We prove that this generalization preserves the desirable theoretical properties of Context Tree Weighting on stationary n-Markov sources, and show empirically that this new technique leads to consistent improvements over Context Tree Weighting as measured on the Calgary Corpus.