Title: Non-parametric policy gradients: A unified treatment of propositional and relational domains
Authors: Kersting, Kristian ×
Driessens, Kurt #
Issue Date: Jul-2008
Host Document: Proceedings of the 25th International Conference on Machine Learning pages:456-463
Conference: International Conference on Machine Learning edition:25 location:Helsinki, Finland date:5-9 July 2008
Abstract: Policy gradient approaches are a powerful instrument
for learning how to interact with
the environment. Existing approaches have
focused on propositional and continuous domains
only. Without extensive feature engineering,
it is difficult – if not impossible –
to apply them within structured domains, in
which e.g. there is a varying number of objects
and relations among them. In this paper,
we describe a non-parametric policy gradient
approach – called NPPG – that overcomes
this limitation. The key idea is to
apply Friedmann’s gradient boosting: policies
are represented as a weighted sum of regression
models grown in an stage-wise optimization.
Employing off-the-shelf regression
learners, NPPG can deal with propositional,
continuous, and relational domains in a unified
way. Our experimental results show that
it can even improve on established results.
ISBN: 978-1-60558-205-4
Publication status: published
KU Leuven publication type: IC
Appears in Collections:Informatics Section
× corresponding author
# (joint) last author

Files in This Item:
File Description Status SizeFormat
2008_icml_kersting.pdfMain article Published 414KbAdobe PDFView/Open


All items in Lirias are protected by copyright, with all rights reserved.