Beygelzimer, Alina , Langford, John , Lifshits, Yuri , Sorkin, Gregory B. and Strehl, Alex (2009) Conditional probability tree estimation analysis and algorithms. In: Uncertainty in artificial intelligence , 18-21 June 2009, Montreal, QC, Canada.
We consider the problem of estimating the conditional probability of a label in time $O(\log n)$, where $n$ 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 $10^6$ labels.
|Item Type:||Conference or Workshop Item (Paper)|
|Additional Information:||© 2009 the Authors|
|Library of Congress subject classification:||Q Science > QA Mathematics > QA75 Electronic computers. Computer science|
|Sets:||Research centres and groups > Management Science Group
Departments > Management
Actions (login required)
|Record administration - authorised staff only|