PAC-Bayes Tree: weighted subtrees with guarantees

Abstract

We present a weighted-majority classification approach over subtrees of a fixed tree, which provably achieves excess-risk of the same order as the best tree-pruning. Furthermore, the computational efficiency of pruning is maintained at both training and testing time despite having to aggregate over an exponential number of subtrees. We believe this is the first subtree aggregation approach with such guarantees.

Publication
In Neural Information Processing Systems 2018
Tin Nguyen
Tin Nguyen
PhD Student

My research interests include scalable and robust Bayesian inference.