Online learning and exploiting relational models in reinforcement learning

Authors:
Tom Croonenborghs;Jan Ramon;Hendrik Blockeel;Maurice Bruynooghe
Affiliations:
K.U. Leuven, Dept. of Computer Science, Leuven;K.U. Leuven, Dept. of Computer Science, Leuven;K.U. Leuven, Dept. of Computer Science, Leuven;K.U. Leuven, Dept. of Computer Science, Leuven
Venue:
IJCAI'07 Proceedings of the 20th international joint conference on Artifical intelligence
Year:
2007

Citing 11
Cited 12

Dyna, an integrated architecture for learning, planning, and reacting

ACM SIGART Bulletin
Prioritized Sweeping: Reinforcement Learning with Less Data and Less Time

Machine Learning
Temporal difference learning and TD-Gammon

Communications of the ACM
Introduction to Reinforcement Learning

Introduction to Reinforcement Learning
A Sparse Sampling Algorithm for Near-Optimal Planning in Large Markov Decision Processes

Machine Learning
KnightCap: A Chess Programm That Learns by Combining TD(lambda) with Game-Tree Search

ICML '98 Proceedings of the Fifteenth International Conference on Machine Learning
Learning relational probability trees

Proceedings of the ninth ACM SIGKDD international conference on Knowledge discovery and data mining
Bellman goes relational

ICML '04 Proceedings of the twenty-first international conference on Machine learning
Learning planning rules in noisy stochastic worlds

AAAI'05 Proceedings of the 20th national conference on Artificial intelligence - Volume 2
Approximate policy iteration with a policy language bias: solving relational Markov decision processes

Journal of Artificial Intelligence Research
A comparison of approaches for learning probability trees

ECML'05 Proceedings of the 16th European conference on Machine Learning

Generalized ordering-search for learning directed probabilistic logical models

Machine Learning
Generalized Ordering-Search for Learning Directed Probabilistic Logical Models

Inductive Logic Programming
Approximate inference for planning in stochastic relational worlds

ICML '09 Proceedings of the 26th Annual International Conference on Machine Learning
Relevance Grounding for Planning in Relational Domains

ECML PKDD '09 Proceedings of the European Conference on Machine Learning and Knowledge Discovery in Databases: Part I
Learning models of relational MDPs using graph kernels

MICAI'07 Proceedings of the artificial intelligence 6th Mexican international conference on Advances in artificial intelligence
Exploration in relational worlds

ECML PKDD'10 Proceedings of the 2010 European conference on Machine learning and knowledge discovery in databases: Part II
Planning with noisy probabilistic relational rules

Journal of Artificial Intelligence Research
Incremental learning of relational action models in noisy environments

ILP'10 Proceedings of the 20th international conference on Inductive logic programming
Probabilistic relational planning with first order decision diagrams

Journal of Artificial Intelligence Research
Active learning of relational action models

ILP'11 Proceedings of the 21st international conference on Inductive Logic Programming
Distributed relational temporal difference learning

Proceedings of the 2013 international conference on Autonomous agents and multi-agent systems
Exploration in relational domains for model-based reinforcement learning

The Journal of Machine Learning Research

Quantified Score

Hi-index	0.00

Visualization

Abstract

In recent years, there has been a growing interest in using rich representations such as relational languages for reinforcement learning. However, while expressive languages have many advantages in terms of generalization and reasoning, extending existing approaches to such a relational setting is a non-trivial problem. In this paper, we present a first step towards the online learning and exploitation of relational models. We propose a representation for the transition and reward function that can be learned online and present a method that exploits thesemodels by augmenting Relational Reinforcement Learning algorithms with planning techniques. The benefits and robustness of our approach are evaluated experimentally.