Context tree weighting

The context tree weighting method (CTW) is a lossless compression and prediction algorithm by Willems, Shtarkov & Tjalkens 1995. The CTW algorithm is among the very few such algorithms that offer both theoretical guarantees and good practical performance (see, e.g. Begleiter, El-Yaniv & Yona 2004). The CTW algorithm is an “ensemble method”, mixing the predictions of many underlying variable order Markov models, where each such model is constructed using zero-order conditional probability estimators.

References

• Willems; Shtarkov; Tjalkens (1995), "The Context-Tree Weighting Method: Basic Properties", IEEE Transactions on Information Theory (IEEE Transactions on Information Theory) 41 (3): 653–664, doi:10.1109/18.382012, https://ieeexplore.ieee.org/document/382012 
• Willems; Shtarkov; Tjalkens (1997), Reflections on "The Context-Tree Weighting Method: Basic Properties", 47, IEEE Information Theory Society Newsletter 
• Begleiter; El-Yaniv; Yona (2004), "On Prediction Using Variable Order Markov Models", Journal of Artificial Intelligence Research (Journal of Artificial Intelligence Research) 22: 385–421, doi:10.1613/jair.1491, https://www.jair.org/index.php/jair/article/view/10394 

External links

• Relevant CTW papers and implementations
• CTW Official Homepage