ABSTRACT
For the purposes of this paper self-improvement will be defined to be the automatic development of problem solving heuristics. The reason for such a narrow definition will be indicated. Precise formulation of the concepts of problem and game will form the basis for the discussion of self-improving programs and an attempt to deal with the question of their adequacies.
The relationship between self improvement and the process of learning will be established in specific cases. The idea of “proof generalization” as a basis of self-improvement will also be discussed. Case studies, where they exist and are understood will be indicated.
- 1.Rich, E., "Artificial Intelligence", McGraw Hill (1983). Google ScholarDigital Library
- 2.Charniak, E. and McDermott, D., "Artificial Intelligence", Addison Wesley (1985). Google ScholarDigital Library
- 3.Nilsson, N., "Searching Problem Solving and Game Playing Trees for Minimum Cost Solutions", Information Processing 68, Vol. 2, p 1556, North Holland (1970).Google Scholar
- 4.Pearl, J., "Heuristics", Addison Wesley (1984).Google Scholar
- 5.Pohl, I., "First Results on the Effect of Error on Heuristic Search", Machine Intelligence 5, p. 219, Edinburgh University Press (1970).Google Scholar
- 6.Gaschnig, J., "Performance Measurement and Analysis of Certain Search Algorithms", Report CMU-CS-79-124, Carnegie Mellon University (1979).Google Scholar
- 7.Gaschnig, J., "A Problem Similarity Approach to Devising Heuristics- First Results" Proc Int Jt Conf on Art Int., Tokyo, p. 301 (1979).Google Scholar
- 8.Banerji, R., "Theory of Problem Solving- A Branch of Artificial Intelligence" Proc Inst of Elec and Electron grs., Vol 70 p. 1428 (1982).Google ScholarCross Ref
- 9.Banerji, R., Artificial Intelligence": A Theoritical Approach", North Holland (1980). Google ScholarDigital Library
- 10.Fikes, R., Hart, P., and Nilsson, N., "Learning and Executing Generalized Robot Plans", Artificial Intelligence, Vol 3, p. 251 (1972).Google ScholarDigital Library
- 11.Dabija, V., Bd. Dr. Petru Groza 41, 76245 Bucaresti, R.S.Romania (Private Communication).Google Scholar
- 12.Ernst, G., "Sufficient Conditions for the Success of GPS", Jour. Assoc. Comp. Mach., Vol 16, p. 517 (1969). Google ScholarDigital Library
- 13.Ernst, G. and Banerji, R., "A Theory for the Complete Mechanization of a GPS-type Problem Solver", Proc. Int. Jt. Conf. on Art. Int., Cambridge, p. 450 (1977).Google Scholar
- 14.Korf, R., "Operator Decomposability: A New Type of Problem Structure", Proc. Nat. Conf. on Art. Int., Washington, D.C., (1983).Google Scholar
- 15.Niizuma, S. and Kitahashi, T., "A Problem-Decomposition Method using Difference or Equivalence Relation Between States" Artificial Intelligence, Vol 25, p. 117 (1985) Google ScholarDigital Library
- 16.Oyen, R., "Mechanical Discovery of Invariances for Problem Solving", Report 1168, Computer Engineering Dept., Case Western Reserve University (1975).Google Scholar
- 17.Ernst G. and Goldstein, M., "Mechanical Discovery of Classes of Problem Solving Strategies", Jour. Assoc. Comp. Mach., Vol 29, p. 1 (1982). Google ScholarDigital Library
- 18.Banerji, R., "The Minmax Criterion in Discreet Games: Economics and AI", Encyclopaedia of Artificial Intelligence, John Wiley and Sons (In Press).Google Scholar
- 19.yon Neuman, J and Morgenstern, O., "Theory of Games and Economic Behavior", Princeton University Press (1947).Google Scholar
- 20.Koffman, E., "Learning Through Pattern Recognition Applied to a Class of Games", IEEE Trans. on Systems, Man and Cybernetics, Vol SSC-4 (March 1968).Google Scholar
- 21.Citrenbaum, R., "Strategic Pattern Recognition: A Strategy for a Class of Games", Pattern Recognition, Vol 4, p. 317 (1972).Google ScholarCross Ref
- 22.Berlekamp, E., Conway, J. and Guy, R., "Winning Ways", Acad. Press (1982).Google Scholar
- 23.Samuel, A., "Some Studies in Machine Learning Using the Game of Checkers", I.B.M. Jour. of Res. and Dev., Vol 3, p. 210 (1958).Google Scholar
- 24.Samuel, A., "some Studies in Machine Learning Using the Game of Checkers II", I.B.M. Jour. of Res. and Dev., 11, p. 601 (1968).Google Scholar
- 25.Banerji, R., "Theory of Problem Solving: An Approach to Artificial Intelligence", Elsevier (1969).Google Scholar
- 26.Nau, D., "Pathology on Game Trees: A Summary of Results" Proc 1st Annual National Conference on Artificial Intelligence, Standford p. 102 (1980).Google Scholar
- 27.Banerji, R., "The Logic of Learning: A Basis for Pattern Recognition and the Improvement of Performance", Advances in Computers (Yovits, Ed.) , Vol 24, p. 177, Academic Press (1985).Google ScholarCross Ref
- 28.Mitchell, T., Utgoff, P and Banerji, R., "Learning by Experimentation: Acquiring and Refining Problem Solving Heuristics" in Machine Learning (Michalski, Carbonell and Mitchell, Eds.), p. 163 Tioga Press (1983).Google Scholar
- 29.Mitchell, T., "Version Spaces- A Candidate Elimination Approach to Rule Learning", Proc. 5th int. Jt. Conf. on Art. Int., Cambridge, p. 305 (1977).Google Scholar
- 30.Plaistead, D., "Using Examples, Case Analysis and Dependency Graphs in Theorem Proving", Proc. 7th int. Conf. on Automated Deduction, (Shostak, Ed.), p. 356, Springer Verlag (1986). Google ScholarDigital Library
- 31.Minton, S., "Constraint-Based Generalization: Learning Game Playing Plans from Single Examples" Proc Amer Assoc for Art. int., Austin, p. 251 (1984).Google Scholar
- 32.King, P., "A Computer Program for Playing Positional Games", M.S. Thesis, Case Western Reserve University (1970).Google Scholar
- 33.Quinlan, J and Hunt, E., "A Formal Deductive Problem Solving System", Jour. Assoc. Comp. Mach., Vol. 15, p. 625 (1968). Google ScholarDigital Library
- 34.Coray, G., "Additive Features in Positional Games", in Computer Oriented Learning Processes (Simon, Ed.), Noordhoff, p. 369 (1976).Google ScholarCross Ref
- 35.Newell, A., Shaw, J. and Simon, H., "Report on a General Problem Solving Program", Proc. Int. Conf. on Inf. Proc. (UNESCO), Paris, p. 256 (1969).Google Scholar
- 36.Banerji, R., "Similarities in Games and their Use in Strategy Construction", Pro~. Symp. on Theory of Computing, Brooklyn Polytechnic Institute (April 1970).Google Scholar
- 37.Grundy, P., "Mathematics and Games", Eureka, Vol. 2, p. 6 (1939).Google Scholar
- 38.Minsky, M. and Papert, S., "Perceptrons", M.I.T. Press (1969).Google Scholar
Index Terms
- Self-improvement in problem-solving
Recommendations
An efficient tabu search for solving the uncapacitated single allocation hub location problem
Given an efficient tabu search the uncapacitated single allocation hub location problem.Given some new tabu rules.Given some new theoretical results for computing the changes in the objective function's value for three types of moves.Improved the best ...
Authoring problem-solving ITS with ASTUS
ITS'10: Proceedings of the 10th international conference on Intelligent Tutoring Systems - Volume Part IIStep-based ITS have been proven successful for well-defined domains, particularly in well-defined tasks, but their success is mitigated by their cost Typically, the main factor behind the cost is the efforts needed to model the task domain Different ...
Comments