Knowledge representation for mathematical discovery: Three experiments in graph theory

This paper describes the nature of mathematical discovery (including concept definition and exploration, example generation, and theorem conjecture and proof), and considers how such an intelligent process can be simulated by a machine. Although the material is drawn primarily from graph theory, the results are immediately relevant to research in mathematical discovery and learning.The thought experiment, a protocol paradigm for the empirical study of mathematical discovery, highlights behavioral objectives for machine simulation. This thought experiment provides an insightful account of the discovery process, motivates a framework for describing mathematical knowledge in terms of object classes, and is a rich source of advice on the design of a system to perform discovery in graph theory. The evaluation criteria for a discovery system, it is argued, must include both a set of behavior to display (behavioral objectives) and a target set of facts to be discovered (factual knowledge).Cues from the thought experiment are used to formulate two hierarchies of representational languages for graphy theory. The first hierarchy is based on the superficial terminology and knowledge of the thought experiment. Generated by formal grammars with set-theoretic semantics, this eminently reasonable approach ultimately fails to meet the factual knowledge criteria. The second hierarchy uses declarative expressions, each of which has a semantic interpretation as a stylized, recursive algorithm that defines a class by generating it correctly and completely. A simple version of one such representation is validated by a successful, implemented system called Graph Theorist (GT) for mathematical research in graph theory. GT generates correct examples, defines and explores new graph theory properties, and conjectures and proves theorems.Several themes run through this paper. The first is the dual goals, behavioral objectives and factural knowledge to be discovered, and the multiplicity of their demands on a discovery system. The second theme is the central role of object classes to knowledge representation. The third is the increased power and flexibility of a constructive (generator) definition over the more traditional predicate (tester) definition. The final theme is the importance of examples and recursion in mathematical knowledge. The results provide important guidance for further research in the simulation of mathematical discovery.

[1]  I. Lakatos,et al.  Proofs and Refutations: Frontmatter , 1976 .

[2]  I. G. McFetridge.  Proofs and Refutations: The Logic of Mathematical Discovery By Imre Lakatos Edited by John Worrall and Elie Zahar Cambridge University Press, 1976, xii + 174 pp., £7.50, £1.95 paper , 1977 .

[3]  Daniel P. Friedman,et al.  A language extension for graph processing and its formal semantics , 1971, Commun. ACM.

[4]  J. A. Bondy,et al.  Graph Theory with Applications , 1978 .

[5]  Hana Galperin.  Succinct representation of graphs , 1982 .

[6]  E O Hovey,et al.  THE NEW YORK ACADEMY OF SCIENCES. , 2022, Science.

[7]  Jaime G. Carbonell,et al.  An Overview of Machine Learning , 1983 .

[8]  Susan L. Epstein.  Knowledge representation in mathematics: a case study in graph theory , 1983 .

[9]  Susan L. Epstein.  Learning and discovery: one system's search for mathematical knowledge , 1988, Comput. Intell..

[10]  Michael Leyton,et al.  A Process-Grammar for Shape , 1988, Artif. Intell..

[11]  Uppaluri S. R. Murty,et al.  Graph Theory with Applications , 1978 .

[12]  Elwood S. Buffa,et al.  Graph Theory with Applications , 1977 .

[13]  Ranan B. Banerji.  Theory of problem solving - an approach to artificial intelligence , 1969, Modern analytic and computational methods in science and mathematics.

[14]  Andrzej Proskurowski.  Recursive Graphs, Recursive Labelings and Shortest Paths , 1981, SIAM J. Comput..

[15]  Jeffrey D. Ullman,et al.  Introduction to Automata Theory, Languages and Computation , 1979 .

[16]  Imre Lakatos,et al.  On the Uses of Rigorous Proof. (Book Reviews: Proofs and Refutations. The Logic of Mathematical Discovery) , 1977 .

[17]  Barry K. Rosen,et al.  Deriving graphs from graphs by applying a production , 1975, Acta Informatica.

[18]  Paul E. Utgoff,et al.  Shift of bias for inductive concept learning , 1984 .

[19]  H. Poincaré.  La valeur de la science , 1905 .

[20]  Philip Michael Dorin.  Aspects of the implementation of sequential graph rewriting systems , 1982 .

[21]  M. Sims.  Empirical and Analytic Discovery in IL , 1987 .

[22]  Douglas B. Lenat,et al.  AM, an artificial intelligence approach to discovery in mathematics as heuristic search , 1976 .

[23]  Mark N. Wegman,et al.  Summarizing graphs by regular expressions , 1983, POPL '83.

[24]  George Polya,et al.  Mathematical discovery : on understanding, learning, and teaching problem solving , 1962 .

[25]  Frank Harary,et al.  Graph Theory , 2016 .

[26]  Mihalis Yannakakis,et al.  Node-and edge-deletion NP-complete problems , 1978, STOC.

[27]  O. Ore.  Theory of Graphs , 1962 .

[28]  Tom M. Mitchell,et al.  MODEL-DIRECTED LEARNING OF PRODUCTION RULES1 , 1978 .

[29]  守屋 悦朗,et al.  J.E.Hopcroft, J.D. Ullman 著, "Introduction to Automata Theory, Languages, and Computation", Addison-Wesley, A5変形版, X+418, \6,670, 1979 , 1980 .

[30]  Robert S. Boyer,et al.  Computational Logic , 1990, ESPRIT Basic Research Series.

[31]  H. Gelernter,et al.  Realization of a geometry theorem proving machine , 1995, IFIP Congress.