KAPLAN@RED.RUTGERS.EDU.UUCP (03/31/87)
PhD Oral Qualifying Examination for Mr. S. Mahadevan Mr. Mahadevan's examination is scheduled for Wednesday, April 1 at 10:30 AM in Hill 423. The examination committee is chaired by T. Mitchell, and includes T. McCarty, J. Mostow, and L. Steinberg. DCS faculty are welcome to attend; graduate students are invited to the public portion of the examination. Mr. Mahadevan's dissertation proposal is abstracted below: LEARNING DECOMPOSITION METHODS TO IMPROVE HIERARCHICAL PROBLEM-SOLVING PERFORMANCE Previous work in machine learning on improving problem-solving performance has usually assumed a state-space or "flat" problem-solving model. However, problem-solvers in complex domains, such as design, usually employ a hierarchical or problem-reduction strategy to avoid the combinatorial explosion of possible operator sequences. Consequently, in order to apply machine learning to complex domains, hierarchical problem-solvers that automatically improve their performance need to designed. One general approach is to design an interactive problem-solver -- a learning apprentice -- that learns from the problem-solving activity of expert users. In this talk we propose a technique, VBL, by which such a system can learn new problem-reduction operators, or decomposition methods, based on a verification of the correctness of example decompositions. We also discuss two important limitations of the VBL technique -- intractability of verification and specificity of generalization -- and propose solutions to them. Finally, we present a formalization of the problem of learning decomposition methods based on viewing actions and problems as binary relations on states. -------