Abstract
We present an approach to build sequential decision making problems which can test the generalization capabilities of classifier
systems. The approach can be applied to any sequential problem defined over a binary domain and it generates a new problemwith bounded sequential difficulty and bounded generalization difficulty. As an example, we applied the approach to generatetwo problems with simple sequential structure, huge number of states (more than a million), and many generalizations. Theseproblems are used to compare a classifier system with effective generalization (XCS) and a learner without generalization(Q-learning). The experimental results confirm what was previously found mainly using single-step problems: also in sequentialproblems with huge state spaces, XCS can generalize effectively by detecting those substructures that are necessary for optimalsequential behavior.
Users
Please
log in to take part in the discussion (add own reviews or comments).