Welcome to the InfoSci Platform
Relational Representations and Traces for Efficient Reinforcement Learning

Relational Representations and Traces for Efficient Reinforcement Learning

Eduardo F. Morales, Julio H. Zaragoza
Copyright: © 2014 |Pages: 26
ISBN13: 9781466646070|ISBN10: 1466646071|EISBN13: 9781466646087
DOI: 10.4018/978-1-4666-4607-0.ch013
Cite Chapter Cite Chapter

MLA

Morales, Eduardo F., and Julio H. Zaragoza. "Relational Representations and Traces for Efficient Reinforcement Learning." Robotics: Concepts, Methodologies, Tools, and Applications, edited by Information Resources Management Association, IGI Global, 2014, pp. 248-273. https://doi.org/10.4018/978-1-4666-4607-0.ch013

APA

Morales, E. F. & Zaragoza, J. H. (2014). Relational Representations and Traces for Efficient Reinforcement Learning. In I. Management Association (Ed.), Robotics: Concepts, Methodologies, Tools, and Applications (pp. 248-273). IGI Global. https://doi.org/10.4018/978-1-4666-4607-0.ch013

Chicago

Morales, Eduardo F., and Julio H. Zaragoza. "Relational Representations and Traces for Efficient Reinforcement Learning." In Robotics: Concepts, Methodologies, Tools, and Applications, edited by Information Resources Management Association, 248-273. Hershey, PA: IGI Global, 2014. https://doi.org/10.4018/978-1-4666-4607-0.ch013

Export Reference

Mendeley
Favorite

Abstract

This chapter introduces an approach for reinforcement learning based on a relational representation that: (i) can be applied over large search spaces, (ii) can incorporate domain knowledge, and (iii) can use previously learned policies on different, but similar, problems. The underlying idea is to represent states as sets of first order relations, actions in terms of those relations, and to learn policies over such generalized representation. It is shown how this representation can produce powerful abstractions and that policies learned over this generalized representation can be directly applied, without any further learning, to other problems that can be characterized by the same set of relations. To accelerate the learning process, we present an extension where traces of the tasks to be learned are provided by the user. These traces are used to select only a small subset of possible actions increasing the convergence of the learning algorithms. The effectiveness of the approach is tested on a flight simulator and on a mobile robot.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.