English

Conditional Probability Tree Estimation Analysis and Algorithms

Machine Learning 2009-06-04 v2 Artificial Intelligence

Abstract

We consider the problem of estimating the conditional probability of a label in time O(logn)O(\log n), where nn is the number of possible labels. We analyze a natural reduction of this problem to a set of binary regression problems organized in a tree structure, proving a regret bound that scales with the depth of the tree. Motivated by this analysis, we propose the first online algorithm which provably constructs a logarithmic depth tree on the set of labels to solve this problem. We test the algorithm empirically, showing that it works succesfully on a dataset with roughly 10610^6 labels.

Keywords

Cite

@article{arxiv.0903.4217,
  title  = {Conditional Probability Tree Estimation Analysis and Algorithms},
  author = {Alina Beygelzimer and John Langford and Yuri Lifshits and Gregory Sorkin and Alex Strehl},
  journal= {arXiv preprint arXiv:0903.4217},
  year   = {2009}
}
R2 v1 2026-06-21T12:44:05.564Z