Notes and Reading - Lecture #18

Lecture #18 extended reinforcement learning to non-deterministic environments.

Notes

Notes are available in two formats. Both are PDF files.

1 slide per page, PDF.
4 slides per page, PDF.

Reading

The material from this lecture isn't covered in any detail in the textbook. However, there are two very good papers which cover this and related material. These are:

C. Boutilier, T. Dean and S. Hanks, Decision-theoretic planning: structural assumptions and computational leverage, Journal of Artificial Intelligence Research, 11, 1-94, 1999.

and

L. P. Kaelbling, M. L. Littman and A. R. Cassandra, Planning and acting in partially observable stochastic domains, Artificial Intelligence, 101, 99-134, 1998.