Abstract
This paper introduces theGenerate, Prune and Prove (GPP) methodology for discovering definitions of mathematical operators. GPP is a task within the IL exploration discovery system. We developed GPP for use in the discovery of mathematical operators with a wider class of representations than was possible with the previous methods by Lenat and by Shen. GPP utilizes thepurpose for which an operator is created to prune the possible definitions. The relevant search spaces are immense and there exists insufficient information for a complete evaluation of the purpose constraint, so it is necessary to perform a partial evaluation of the purpose (i.e., pruning) constraint. The constraint is first transformed so that it is operational with respect to the partial information, and then it is applied to examples in order to test the generated candidates for an operator's definition. In the GPP process, once a candidate definition survives this empirical prune, it is passed on to a theorem prover for formal verification. In this paper, we describe the application of this methodology to the (re)discovery of the definition of multiplication for Conway numbers, a discovery which is difficult for human mathematicians. We successfully model this discovery process utilizing information which was reasonably available at the time of Conway's original discovery. As part of this discovery process, we reduce the size of the search space from a computationally intractable size to 3468 elements.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
S. Amarel, Program synthesis as a theory formation task — problem representations and solution methods, in:Machine Learning: An Artificial Intelligence Approach, vol. 2, eds. R.S. Michalski, J.G. Carbonell and T.M. Mitchell (Morgan Kaufmann, San Mateo, CA, 1986) pp. 499–569.
J. Backus, Can programming be liberated from the von Neumann style: A functional style and its algebra of programs, Commun. ACM (August 1978).
A. Barr and E.A. Feigenbaum,The Handbook of Artificial Intelligence, vol. 2 (William Kaufmann, Los Altos, Ca, 1982).
R. Bhaskar and N. Nigam, Qualitative physics using dimensional analysis, Art. Int. J. 45 (1990) 73–111.
D. Bjorner, A.P. Ershov and N.D. Jones (eds.),Workshop on Partial Evaluation and Mixed Computation, Lecture Notes in Computer Science (Springer, 1987).
W.W. Bledsoe, Using examples to generate instantiations of set variables, in:Proc. of IJCAI-83, Karlsruhe, Germany (1983) pp. 892–901.
J.L. Bresina, REAPPR — An expert system shell for planning, Technical Report LCSR-TR-119, Laboratory for Computer Science Research, Rutgers University (February, 1988).
A. Bundy,The Computer Modelling of Mathematical Reasoning (Academic Press, Orlando, FL, 1983).
J.G. Carbonell and Y. Gil, Learning by experimentation: the operator refinement method, in:Machine Learning: An Artificial Intelligence Approach, vol. 3, eds. Y. Kodratoff and R.S. Michalski (Morgan Kaufmann, San Mateo, CA, 1990) pp. 191–213.
J.H. Conway,On Numbers and Games (Academic Press, Orlando, FL, 1976).
S.L. Epstein, On the discovery of mathematical theorems, in:Proc. of IJCAI-87, Milan (Morgan Kaufmann, 1987) pp. 194–197.
H. Gonshor,An Introduction to the Theory of Surreal Numbers (Cambridge University Press, New York, 1986).
H. Hirsh, Conditional operationality and explanation-based generalization, in:Machine Learning: An Artificial Intelligence Approach, vol. 3, eds. Y. Kodratoff and R.S. Michalski (Morgan Kaufmann, San Mateo, CA, 1990) pp. 383–395.
S.T. Kedar-Cabelli, Formulating concepts and analogies according to purpose, Ph.D. Thesis, Rutgers University (May 1988).
R.M. Keller, Defining operationalization for explanation-based learning, in:Proc. of AAAI-87, Seattle, WA (Morgan Kaufmann, 1987) pp. 482–487.
M.M. Kokar, Determining arguments of invariant functional descriptions, Machine Learning (1986) 403–422.
D.E. Knuth,Surreal Numbers (Addison-Wesley, 1974).
I. Lakatos,Proofs and Refutations. The Logic of Mathematical Discovery, eds. Worrall and E. Zahar (Cambridge University Press, Cambridge, 1976).
D. Lenat and J.S. Brown, Why AM and Eurisko appear to work, Art. Int. J. 23 (1984) 269–294.
D. Lenat, Automatic theory formation in mathematics, in:Proc. of IJCAI-77, Cambridge, MA (1977) pp. 833–842.
D. Lenat, EURISKO: A program that learns new heuristics and domain concepts. The nature of heuristics III: Program design and results, Art. Int. J. 21 (March, 1983).
M.R. Lowry, Symmetry as bias: Rediscovering special relativity, in:Proc. of AAAI-92, San Jose, CA (AAAI Press, 1992) pp. 56–62.
S.N. Minton, J.G. Carbonell, C.A. Knoblock, D.R. Kuokka, O. Etzioni and Y. Gil, Explanation-based learning: A problem solving perspective, Art. Int. J. 40 (1989) 63–118.
T.M. Mitchell, R.M. Keller and S.T. Kedar-Cabelli, Explanation-based generalization: a unifying view, Machine Learning 1 (1986) 47–80.
J. Mostow, Searching for operational concept descriptions in BAR, MetaLEX, and EBG, in:Proc. Int. Machine Learning Workshop-87, University of California at Irvine, Irvine, CA (Morgan Kaufmann, 1987) pp. 376–382.
N.J. Nilsson,Problem-solving Methods in Artificial Intelligence (McGraw-Hill, New York, 1971).
W. Shen, Functional transformations in AI discovery systems, Art. Int. J. 41 (1990) 257–272.
M.H. Sims and J.L. Bresina, Discovering mathematical operator definitions,Proc. 6th Int. Workshop on Machine Learning, Cornell University (1989) pp. 308–313.
M.H. Sims, IL: an artificial intelligence approach to theory formation in mathematics, Ph.D. Thesis, Rutgers University (May 1990).
H. Gelernter, Realization of a geometry-theorem proving machine, in:Computers and Thought, eds. E. Feigenbaum and Feldman (McGraw-Hill, New York, 1963) pp. 134–152.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Sims, M.H., Bresina, J.L. Purposive discovery of operators. Ann Math Artif Intell 6, 317–343 (1992). https://doi.org/10.1007/BF01535524
Issue Date:
DOI: https://doi.org/10.1007/BF01535524