Fewer Epistemological Challenges for Connectionism

  • Artur S. d’Avila Garcez
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3526)

Abstract

Seventeen years ago, John McCarthy wrote the note Epistemological challenges for connectionism as a response to Paul Smolensky’s paper On the proper treatment of connectionism. I will discuss the extent to which the four key challenges put forward by McCarthy have been solved, and what are the new challenges ahead. I argue that there are fewer epistemological challenges for connectionism, but progress has been slow. Nevertheless, there is now strong indication that neural-symbolic integration can provide effective systems of expressive reasoning and robust learning due to the recent developments in the field.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Andrews, R., Diederich, J., Tickle, A.B.: A survey and critique of techniques for extracting rules from trained artificial neural networks. Knowledge-based Systems 8(6), 373–389 (1995)CrossRefGoogle Scholar
  2. 2.
    Bader, S., d’Avila Garcez, A.S., Hitzler, P.: Computing first order logic programs by fibring artificial neural networks. In: Proceedings of International FLAIRS Conference, Florida, USA. AAAI Press, Menlo Park (2005)Google Scholar
  3. 3.
    Bologna, G.: Is it worth generating rules from neural network ensembles? In: d’Avila Garcez, A.S., Gabbay, D., Hölldobler, S., Taylor, J.G. (eds.) Journal of Applied Logic, Special issue on Neural-Symbolic Integration, vol. 2(3), pp. 325–348 (2004)Google Scholar
  4. 4.
    Cloete, I., Zurada, J.M. (eds.): Knowledge-Based Neurocomputing. MIT Press, Cambridge (2000)Google Scholar
  5. 5.
    d’Avila Garcez, A.S., Broda, K., Gabbay, D.M.: Symbolic knowledge extraction from trained neural networks: A sound approach. Artificial Intelligence 125, 155–207 (2001)MATHMathSciNetCrossRefGoogle Scholar
  6. 6.
    d’Avila Garcez, A.S., Broda, K., Gabbay, D.M.: Neural-Symbolic Learning Systems: Foundations and Applications. Perspectives in Neural Computing. Springer, Heidelberg (2002)Google Scholar
  7. 7.
    d’Avila Garcez, A.S., Gabbay, D.M.: Fibring neural networks. In: Proceedings of 19th National Conference on Artificial Intelligence AAAI 2004, San Jose, California, USA, pp. 342–347. AAAI Press, Menlo Park (2004)Google Scholar
  8. 8.
    d’Avila Garcez, A.S., Lamb, L.C.: Reasoning about time and knowledge in neural-symbolic learning systems. In: Thrun, S., Saul, L., Schoelkopf, B. (eds.) Proceedings of the NIPS 2003 Conference, Vancouver, Canada. Advances in Neural Information Processing Systems, vol. 16, pp. 921–928. MIT Press, Cambridge (2004)Google Scholar
  9. 9.
    d’Avila Garcez, A.S., Lamb, L.C., Broda, K., Gabbay, D.M.: Applying connectionist modal logics to distributed knowledge representation problems. International Journal of Artificial Intelligence Tools 13(1), 115–139 (2004)CrossRefGoogle Scholar
  10. 10.
    Gabbay, D.M.: Elementary Logics: a Procedural Perspective. Prentice-Hall, Englewood Cliffs (1998)Google Scholar
  11. 11.
    Gabbay, D.M.: Fibring Logics. Oxford University Press, Oxford (1999)MATHGoogle Scholar
  12. 12.
    Haykin, S.: Neural Networks: A Comprehensive Foundation. Prentice-Hall, Englewood Cliffs (1999)MATHGoogle Scholar
  13. 13.
    Hitzler, P., Hölldobler, S., Seda, A.: Logic programs and connectionist networks. Journal of Applied Logic, Special issue on Neural-Symbolic Integration 2(3), 245–272 (2004); d’Avila Garcez, A.S., Gabbay, D., Hölldobler, S., Taylor, J.G. (eds.)Google Scholar
  14. 14.
    Hölldobler, S.: Automated inferencing and connectionist models. Postdoctoral Thesis, Intellektik, Informatik, TH Darmstadt (1993)Google Scholar
  15. 15.
    Hölldobler, S., Kalinke, Y., Storr, H.P.: Approximating the semantics of logic programs by recurrent neural networks. In: Kurfess, F. (ed.) Applied Intelligence Journal, Special Issue on Neural Networks and Structured Knowledge, vol. 11(1), pp. 45–58 (1999)Google Scholar
  16. 16.
    Lloyd, J.W.: Logic for Learning: Learning Comprehensible Theories from Structured Data. Springer, Heidelberg (2003)MATHGoogle Scholar
  17. 17.
    McCarthy, J.: Epistemological challenges for connectionism. Behavior and Brain Sciences 11(1), 44 (1988)CrossRefGoogle Scholar
  18. 18.
    Rumelhart, D.E., Hinton, G.E., Williams, R.J.: Learning internal representations by error propagation. In: Rumelhart, D.E., McClelland, J.L. (eds.) Parallel Distributed Processing: Explorations in the Microstructure of Cognition, vol. 1, pp. 318–362. MIT Press, Cambridge (1986)Google Scholar
  19. 19.
    Setiono, R.: Extracting rules from neural networks by pruning and hidden-unit splitting. Neural Computation 9, 205–225 (1997)MATHCrossRefGoogle Scholar
  20. 20.
    Shastri, L.: Advances in SHRUTI: a neurally motivated model of relational knowledge representation and rapid inference using temporal synchrony. In: Kurfess, F. (ed.) Applied Intelligence Journal, Special Issue on Neural Networks and Structured Knowledge, vol. 11, pp. 79–108 (1999)Google Scholar
  21. 21.
    Shavlik, J.W.: An overview of research at Wisconsin on knowledge-based neural networks. In: Proceedings of the International Conference on Neural Networks ICNN 1996, Washington, DC, pp. 65–69 (1996)Google Scholar
  22. 22.
    Smolensky, P.: On the proper treatment of connectionism. Behavioral and Brain Sciences 11, 1–74 (1988)CrossRefGoogle Scholar
  23. 23.
    Sun, R.: Robust reasoning: integrating rule-based and similarity-based reasoning. Artificial Intelligence 75(2), 241–296 (1995)CrossRefGoogle Scholar
  24. 24.
    Thrun, S.B.: Extracting provably correct rules from artificial neural networks. Technical report, Institut fur Informatik, Universitat Bonn (1994)Google Scholar
  25. 25.
    Valiant, L.G.: Three problems in computer science. Journal of the ACM 50(1), 96–99 (2003)CrossRefGoogle Scholar
  26. 26.
    Vardi, M.Y.: Why is modal logic so robustly decidable. In: Immerman, N., Kolaitis, P. (eds.) Descriptive Complexity and Finite Models. Discrete Mathematics and Theoretical Computer Science, vol. 31, pp. 149–184 (1997)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2005

Authors and Affiliations

  • Artur S. d’Avila Garcez
    • 1
  1. 1.Dept. of Computing School of InformaticsCity University LondonUK

Personalised recommendations