You are here

Lecture Notes in Computer ScienceMachine Learning and Knowledge Discovery in Databases

TitleLecture Notes in Computer ScienceMachine Learning and Knowledge Discovery in Databases
Publication TypeBook Chapter
Year of Publication2008
AuthorsDietterich, T. G., and M. Wynkoop
Secondary AuthorsDaelemans, W., B. Goethals, and K. Morik
Book TitleMachine Learning and Knowledge Discovery in Databases
Series TitleLecture Notes in Computer Science
Pagination597 - 612
Date Published2008
PublisherSpringer Berlin Heidelberg
CityBerlin, Heidelberg
ISBN Number978-3-540-87481-2

This paper addresses the problem of learning dynamic Bayesian network (DBN) models to support reinforcement learning. It focuses on learning regression tree (context-specific dependence) models of the conditional probability distributions of the DBNs. Existing algorithms rely on standard regression tree learning methods (both propositional and relational). However, such methods presume that the stochasticity in the domain can be modeled as a deterministic function with additive noise. This is inappropriate for many RL domains, where the stochasticity takes the form of stochastic choice over deterministic functions. This paper introduces a regression tree algorithm in which each leaf node is modeled as a finite mixture of deterministic functions. This mixture is approximated via a greedy set cover. Experiments on three challenging RL domains show that this approach finds trees that are more accurate and that are more likely to correctly identify the conditional dependencies in the DBNs based on small samples.