Webpage for:
Multiplicative Forests for Continuous-Time Processes
Jeremy C Weiss, Sriraam Natarajan, and David Page
PDF
Citation
Abstract:
Learning temporal dependencies between variables over continuous time is an important and challenging task. Continuous-time Bayesian networks effectively model such processes but are limited by the number of conditional intensity matrices, which grows exponentially in the number of parents per variable. We develop a partition-based representation using regression trees and forests whose parameter spaces grow linearly in the number of node splits. Using a multiplicative assumption we show how to update the forest likelihood in closed form, producing efficient model updates. Our results show multiplicative forests can be learned from few temporal trajectories with large gains in performance and scalability.
mfCTBN code.
Here is some research code for doing multiplicative forest continuous-time Bayesian network learning. Read the readme.txt to get started.
Code ball
NIPS 2012 Supplement.
Below are the extra figures omitted from the proceedings. It includes pseudocode, additional experimental figures, and forest models.
Supplement
Jeremy Weiss
[webpage]
cs.wisc.edu: jcweiss@
Department of Computer Science, Medicine
Advisor: David Page