No Access

REINFORCEMENT LEARNING WITH GOAL-DIRECTED ELIGIBILITY TRACES

M. ANDRECUT

Department of Physics, University of Lethbridge, 4401 University Drive, Lethbridge, AB, T1K 3M4, Canada

Search for more papers by this author

and

M. K. ALI

Department of Physics, University of Lethbridge, 4401 University Drive, Lethbridge, AB, T1K 3M4, Canada

Search for more papers by this author

https://doi.org/10.1142/S0129183104006662Cited by:1 (Source: Crossref)

Abstract

The eligibility trace is the most important mechanism used so far in reinforcement learning to handle delayed reward. Here, we introduce a new kind of eligibility trace, the goal-directed trace, and show that it results in more reliable learning than the conventional trace. In addition, we also propose a new efficient algorithm for solving the goal-directed reinforcement learning problem.

Keywords:

You currently do not have access to the full text article.
Recommend the journal to your library today!

References

L. P. Kaelbling, M. L. Littman and A. W. Moore, J. Artificial Intelligence Res. 4, 237 (1996). Crossref, Web of Science, Google Scholar
R. S. Sutton and A. G. Barto , Reinforcement Learning: An Introduction ( MIT Press , Cambridge , 1998 ) . Google Scholar
D. P. Bertsekas and J. N. Tsitsiklis , Neuro-Dynamic Programming ( Athena Scientific , Belmont , 1996 ) . Google Scholar
R. S. Sutton, Machine Learning 3, 9 (1988). Google Scholar
S. P. Singh and R. S. Sutton, Machine Learning 22, 123 (1996). Web of Science, Google Scholar
T. Jaakkola, M. I. Jordan and S. P. Singh, Neural Computation 6, 1185 (1994). Crossref, Web of Science, Google Scholar
S. Koenig and R. G. Simmons, Machine Learning 22, 227 (1996). Web of Science, Google Scholar