AN APPRENTICE-BASED APPROACH TO KNOWLEDGE ACQUISITION

We explore here the feasibility of learning apprentice programs: inter active knowledge-based assistants that learn by observing and analyzin g the problem-solving steps of their users. In particular, we describe a learning apprentice for digital circuit design, called LEAP. LEAP l earns feasible ways of decomposing circuit modules into submodules, as well as the recommended method when there are competing feasible deco mpositions. VBL is an explanation-based learning technique used in LEA P to infer problem-reduction operators for decomposing circuit modules . PED is a general extension of explanation-based learning to incomple te domain theories containing determinations. PED is used in LEAP to l earn control rules for ranking alternative decompositions as well as t o extend LEAP's partial theory of circuit cost. An experimental study shows that by using this approach LEAP can learn a significant subset of a manually created knowledge base for boolean circuit design. The e xperimental study also reveals some limitations of LEAP, and more gene rally suggests directions for further research in building effective l earning apprentice systems.