Lessons from Theory Revision Applied to Constructive Induction

Abstract

Two challenges of integrating theory knowledge and inductive learning are: 1) A representation language appropriate for a coarse theory may be inappropriate for an improved theory. While the original representation may concisely express the initial theory, a more accurate theory forced to use that same representation may be bulky, cumbersome, and difficult to reach. 2) A theory structure suitable for a coarse domain theory may be insufficient for a fine-tuned theory. Systems that produce only small, local changes to a theory have limited value for accomplishing complex structural alterations that may be required. An analysis of various theory revision systems and theory-guided learning systems reveals specific strengths and weaknesses in terms of these two challenges. Designed to capture the underlying qualities of each system, and to implement flexibility of representation and structure, a new system uses theory-guided constructive induction. Experiments in three domains show improvement over previous systems.