CONSTRUCTIVE REINFORCEMENT LEARNING


Authors

José Hernández-Orallo

Abstract

This paper presents an operative measure of reinforcement for constructive learning methods, i.e., eager learning methods using highly expressible (or universal) representation languages. These evaluation tools allow a further insight in the study of the growth of knowledge, theory revision and abduction. The final ap-proach is based on an apportionment of credit wrt. the ‘course’ that the evidence makes through the learnt theory. Our measure of reinforcement is shown to be justified by cross-validation and by the connection with other successful evaluation criteria, like the MDL principle. Finally, the relation with the classical view of reinforcement is studied, where the actions of an intelligent system can be rewarded or penalised, and we discuss whether this should affect our distribution of reinforcement. The most important result of this paper is that the way we distribute reinforcement into knowledge results in a rated ontology, instead of a single prior distribution. Therefore, this detailed information can be exploited for guiding the space search of in-ductive learning algorithms. Likewise, knowledge revision may be done to the part of the theory which is not justified by the evidence.

Keywords

Reinforcement Learning; Theory Evaluation; Incremental Learning; Ontology; Apportionment of Credit; Abduction; Induction; MDL principle; Knowledge Acquisition and Revision; ILP; Philosophy of Science