Axioms for Rational Reinforcement Learning

Details Axioms for Rational Reinforcement Learning Axioms for Rational Reinforcement Learning

2011-07-27

arxiv.org/abs/1107.5520v1

Proc. 22nd International Conf. on Algorithmic Learning Theory (ALT-2011) pages 338-352

Computer Science

Learning

16 LaTeX pages

Scientific paper

We provide a formal, simple and intuitive theory of rational decision making including sequential decisions that affect the environment. The theory has a geometric flavor, which makes the arguments easy to visualize and understand. Our theory is for complete decision makers, which means that they have a complete set of preferences. Our main result shows that a complete rational decision maker implicitly has a probabilistic model of the environment. We have a countable version of this result that brings light on the issue of countable vs finite additivity by showing how it depends on the geometry of the space which we have preferences over. This is achieved through fruitfully connecting rationality with the Hahn-Banach Theorem. The theory presented here can be viewed as a formalization and extension of the betting odds approach to probability of Ramsey and De Finetti.

Affiliated with

Also associated with

No associations

Rating

Axioms for Rational Reinforcement Learning does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with Axioms for Rational Reinforcement Learning, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Axioms for Rational Reinforcement Learning will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-113625