Gradient-based boosting for Statistical Relational Learning: The Relational Dependency Network Case

DMCA

Gradient-based boosting for Statistical Relational Learning: The Relational Dependency Network Case (2011)

Cached

Download Links

by Sriraam Natarajan , Tushar Khot , Kristian Kersting , Bernd Gutmann , Jude Shavlik

Citations:

39 - 17 self

BibTeX

@MISC{Natarajan11gradient-basedboosting,
    author = {Sriraam Natarajan and Tushar Khot and Kristian Kersting and Bernd Gutmann and Jude Shavlik},
    title = {Gradient-based boosting for Statistical Relational Learning: The Relational Dependency Network Case},
    year = {2011}
}

OpenURL

Abstract

Abstract. Dependency networks approximate a joint probability distribution over multiple random variables as a product of conditional distributions. Relational Dependency Networks (RDNs) are graphical models that extend dependency networks to relational domains. This higher expressivity, however, comes at the expense of a more complex modelselection problem: an unbounded number of relational abstraction levels might need to be explored. Whereas current learning approaches for RDNs learn a single probability tree per random variable, we propose to turn the problem into a series of relational function-approximation problems using gradient-based boosting. In doing so, one can easily induce highly complex features over several iterations and in turn estimate quickly a very expressive model. Our experimental results in several different data sets show that this boosting method results in efficient learning of RDNs when compared to state-of-the-art statistical relational learning approaches. 1

Keyphrases

gradient-based boosting relational dependency network case statistical relational learning dependency network whereas current learning approach multiple random variable unbounded number relational function-approximation problem relational dependency network expressive model joint probability distribution relational domain efficient learning turn estimate graphical model complex modelselection problem method result complex feature experimental result several iteration conditional distribution single probability relational abstraction level several different data set state-of-the-art statistical relational learning approach

DMCA