Advertisement

A Recursive Neural Network for Reflexive Reasoning

  • Steffen Hölldobler
  • Yvonne Kalinke
  • Jörg Wunderlich
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1778)

Abstract

We formally specify a connectionist system for generating the least model of a datalogic program which uses linear time and space. The system is shown to be sound and complete if only unary relation symbols are involved and complete but unsound otherwise. For the latter case a criteria is defined which guarantees correctness. Finally, we compare our system to the forward reasoning version of Shruti.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Apt, K.R., Van Emden, M.H.: Contributions to the theory of logic programming. Journal of the ACM 29, 841–862 (1982)zbMATHCrossRefGoogle Scholar
  2. 2.
    Beringer, A., Hölldobler, S.: On the adequateness of the connection method. In: Proceedings of the AAAI National Conference on Artificial Intelligence, pp. 9–14 (1993)Google Scholar
  3. 3.
    Bibel, W.: On matrices with connections. Journal of the ACM 28, 633–645 (1981)zbMATHCrossRefMathSciNetGoogle Scholar
  4. 4.
    Bibel, W.: Advanced topics in automated deduction. In: Nossum, R.T. (ed.) ACAI 1987. LNCS, vol. 345, pp. 41–59. Springer, Heidelberg (1988)Google Scholar
  5. 5.
    Bibel, W.: Deduction. Academic Press, London (1993)Google Scholar
  6. 6.
    d’Avila Garcez, A.S., Zaverucha, G., de Carvalho, L.A.V.: Logic programming and inductive learning in artificial neural networks. In: Herrmann, F.C., Reine, F., Strohmaier, A. (eds.) Knowledge Representation in Neural Networks, Berlin, pp. 33–46. Logos (1997)Google Scholar
  7. 7.
    Dowling, W.F., Gallier, J.H.: Linear-time algorithms for testing the satisfiability of propositional Horn formulae. Journal of Logic Programming 1(3), 267–284 (1984)zbMATHCrossRefMathSciNetGoogle Scholar
  8. 8.
    Elcock, E.W., Hoddinott, P.: Comments on Kornfeld’s equality for Prolog: E-unification as a mechanism for argumenting the prolog search strategy. In: Proceedings of the AAAI National Conference on Artificial Intelligence, pp. 766–774 (1986)Google Scholar
  9. 9.
    Feldman, J.A.: Memory and change in connection networks. Technical Report TR96, Computer Science Department, University of Rochester (1981)Google Scholar
  10. 10.
    Feldman, J.A., Ballard, D.H.: Connectionist models and their properties. Cognitive Science 6(3), 205–254 (1982)CrossRefGoogle Scholar
  11. 11.
    Fitting, M.: Metric methods – three examples and a theorem. Journal of Logic Programming 21(3), 113–127 (1994)zbMATHMathSciNetGoogle Scholar
  12. 12.
    Fodor, J.A., Pylyshyn, Z.W.: Connectionism and cognitive architecture: A critical analysis. In: Pinker, Mehler (eds.) Connections and Symbols, pp. 3–71. MIT Press, Cambridge (1988)Google Scholar
  13. 13.
    Funahashi, K.-I.: On the approximate realization of continuous mappings by neural networks. Neural Networks 2, 183–192 (1989)CrossRefGoogle Scholar
  14. 14.
    Hölldobler, S.: A structured connectionist unification algorithm. In: Proceedings of the AAAI National Conference on Artificial Intelligence, pp. 587–593 (1990)Google Scholar
  15. 15.
    Hölldobler, S., Kalinke, Y.: Towards a massively parallel computational model for logic programming. In: Proceedings of the ECAI 1994 Workshop on Combining Symbolic and Connectionist Processing, ECCAI, pp. 68–77 (1994)Google Scholar
  16. 16.
    Hölldobler, S., Kalinke, Y., Lehmann, H.: Designing a counter: Another case study of dynamics and activation landscapes in recurrent networks. In: Brewka, G., Habel, C., Nebel, B. (eds.) KI 1997. LNCS, vol. 1303, pp. 313–324. Springer, Heidelberg (1997)Google Scholar
  17. 17.
    Hölldobler, S., Kalinke, Y., Störr, H.-P.: Recurrent neural networks to approximate the semantics of acceptable logic programs. In: Antoniou, G., Slaney, J.K. (eds.) Canadian AI 1998. LNCS (LNAI), vol. 1502, Springer, Heidelberg (1998)CrossRefGoogle Scholar
  18. 18.
    Hölldobler, S., Kalinke, Y., Störr, H.-P.: Approximating the semantics of logic programs by recurrent neural networks. Applied Intelligence 11, 45–59 (1999)CrossRefGoogle Scholar
  19. 19.
    Johnson-Laird, P.N., Byrne, R.M.J.: Deduction. Lawrence Erlbaum Associates, Hove and London, UK (1991)Google Scholar
  20. 20.
    Kalinke, Y.: Using connectionist term representation for first–order deduction – a critical view. In: Maire, F., Hayward, R., Diederich, J. (eds.) Connectionist Systems for Knowledge Representation Deduction, Queensland University of Technology, CADE–14 Workshop, Townsville, Australia (1997)Google Scholar
  21. 21.
    Kalinke, Y., Lehmann, H.: Computations in recurrent neural networks: From counters to iterated function systems. In: Antoniou, G., Slaney, J.K. (eds.) Canadian AI 1998. LNCS (LNAI), vol. 1502. Springer, Heidelberg (1998)CrossRefGoogle Scholar
  22. 22.
    Karp, R.M., Ramachandran, V.: Parallel algorithms for shared-memory machines. In: van Leeuwen, J. (ed.) Handbook of Theoretical Computer Science, ch.17, pp. 869–941. Elsevier Science Publishers B.V, New York (1990)Google Scholar
  23. 23.
    Lloyd, J.W.: Foundations of Logic Programming. Springer, Heidelberg (1987)zbMATHGoogle Scholar
  24. 24.
    McCulloch, W.S., Pitts, W.: A logical calculus and the ideas immanent in nervous activity. Bulletin of Mathematical Biophysics 5, 115–133 (1943)zbMATHCrossRefMathSciNetGoogle Scholar
  25. 25.
    Newell, A.: Physical symbol systems. Cognitive Science 4, 135–183 (1980)CrossRefGoogle Scholar
  26. 26.
    Pinkas, G.: Symmetric neural networks and logic satisfiability. Neural Computation 3, 282–291 (1991)CrossRefGoogle Scholar
  27. 27.
    Plate, T.A.: Holographic reduced representations. In: Proceedings of the International Joint Conference on Artificial Intelligence, pp. 30–35 (1991)Google Scholar
  28. 28.
    Pollack, J.B.: Recursive auto-associative memory: Devising compositional distributed representations. In: Proceedings of the Annual Conference of the Cognitive Science Society, pp. 33–39 (1988)Google Scholar
  29. 29.
    Shastri, L., Ajjanagadde, V.: From associations to systematic reasoning: A connectionist representation of rules, variables and dynamic bindings using temporal synchrony. Behavioural and Brain Sciences 16(3), 417–494 (1993)CrossRefGoogle Scholar
  30. 30.
    Sperduti, A.: Labeling RAAM. Technical Report TR-93-029, International Computer Science Institute, Berkeley, CA (1993)Google Scholar
  31. 31.
    Towell, G.G., Shavlik, J.W.: Extracting refined rules from knowledge–based neural networks. Machine Learning 131, 71–101 (1993)Google Scholar
  32. 32.
    Ullman, J.D.: Principles of Database Systems. Computer Science Press, Rockville (1985)Google Scholar
  33. 33.
    Wunderlich, J.: Erweiterung des RNN–Modells um SHRUTI–Konzepte: Vom aussagenlogischen zum Schlieβen uber prädikatenlogischen Programmen. Master’s thesis, TU Dresden, Fakultät Informatik (1998)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2000

Authors and Affiliations

  • Steffen Hölldobler
    • 1
  • Yvonne Kalinke
    • 2
  • Jörg Wunderlich
    • 3
  1. 1.Dresden University of TechnologyDresdenGermany
  2. 2.Queensland University of TechnologyBrisbaneAustralia
  3. 3.Neurotec Hochtechnologie GmbHFriedrichshafenGermany

Personalised recommendations