CMU Artificial Intelligence Repository
Home INFO Search FAQs Repository Root

RWM: Refinement With Macros

areas/learning/systems/rwm/
RWM (Refinement With Macros) is an interactive system for learning problem solving strategies. RWM incorporates two techniques, namely Refinement and Macro Generation. The input to the RWM is the definition of a problem, the output is a strategy for solving the given problem. A strategy is a sequence of subproblems which are easier then the original problem, and when solved in the the given order it yields a solution for the original problem. 
Origin:   

   FTP repositories for machine learning include
      ftp.ics.uci.edu:/pub/
      cs.utexas.edu:/pub/mooney/
      ftp.gmd.de:/gmd/mlt/ML-Program-Library/
      ftp.gmd.de:/MachineLearning/
Version: 17-NOV-92 Requires: Common Lisp Ports: Tested under Lucid CL. CD-ROM: Prime Time Freeware for AI, Issue 1-1 Author(s): H. Altay Guvenir Keywords: Authors!Guvenir, Lisp!Code, Machine Learning!Macro Generation, Machine Learning!Refinement, RWM References: H. A. Guvenir and G. W. Ernst, "Learning Problem Solving Strategies Using Refinement and Macro Generation", Artificial Intelligence, Vol. 44, No. 1-2, July, 1990. pp. 209-243. H. A. Guvenir and V. Akman, "Problem Representation for Refinement", Mind and Machines, Vol. 2, No. 3, August 1992. pp 267-282. H. A. Guvenir and G. W. Ernst, "A Method for Learning Problem Solving Strategies", Proceedings of the AAAI Spring Symposium Series, Stanford University, March 1988. pp 31-35.
Last Web update on Mon Feb 13 10:24:34 1995
AI.Repository@cs.cmu.edu