Machine Learning

, Volume 43, Issue 1, pp 7–52

Relational Reinforcement Learning

  • Sašo Džeroski
  • Luc De Raedt
  • Kurt Driessens

DOI: 10.1023/A:1007694015589

Cite this article as:
Džeroski, S., De Raedt, L. & Driessens, K. Machine Learning (2001) 43: 7. doi:10.1023/A:1007694015589


Relational reinforcement learning is presented, a learning technique that combines reinforcement learning with relational learning or inductive logic programming. Due to the use of a more expressive representation language to represent states, actions and Q-functions, relational reinforcement learning can be potentially applied to a new range of learning tasks. One such task that we investigate is planning in the blocks world, where it is assumed that the effects of the actions are unknown to the agent and the agent has to learn a policy. Within this simple domain we show that relational reinforcement learning solves some existing problems with reinforcement learning. In particular, relational reinforcement learning allows us to employ structural representations, to abstract from specific goals pursued and to exploit the results of previous learning phases when addressing new (more complex) situations.

reinforcement learning inductive logic programming planning 
Download to read the full article text

Copyright information

© Kluwer Academic Publishers 2001

Authors and Affiliations

  • Sašo Džeroski
    • 1
  • Luc De Raedt
    • 2
  • Kurt Driessens
    • 3
  1. 1.Department of Intelligent SystemsJožef Stefan InstituteLjubljanaSlovenia
  2. 2.Institüt für InformatikAlbert-Lüdwigs-Universität Freiburg, Georges KöhlerFreiburgGermany
  3. 3.Department of Computer ScienceKatholieke Universiteit LeuvenHeverleeBelgium

Personalised recommendations