Uncertainty and Trust Estimation in Incrementally Learning Function Approximation

  • Andreas Buschermöhle
  • Jan Schoenke
  • Werner Brockmann
Part of the Communications in Computer and Information Science book series (CCIS, volume 297)


Incremental learning gets increasingly important to cope with systems of high complexity or to adapt to changing environmental conditions. But to assure safety, the process of incremental learning must be supervised so that no knowledge learned incorrectly or under uncertain conditions influences the system in a contra-productive way. Hence we consider two principles to estimate different kinds of uncertainty, or in other words the trustworthiness, of an incrementally learning system. They are investigated principally for a simplified scenario that explicitly covers all different kinds of uncertainties. Finally, a combined measure to reflect all uncertainties of an incrementally learning system is presented.


Uncertainty Modeling Incremental Learning Function Approximation Trust Management 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Brockmann, W., Buschermöhle, A., Hülsmann, J.: A Generic Concept to Increase the Robustness of Embedded Systems by Trust Management. In: Proc. Int. Conf. on Systems, Man and Cybernetics, pp. 2037–2044. IEEE Press, Istanbul (2010)Google Scholar
  2. 2.
    Buschermoehle, A., Schoenke, J., Brockmann, W.: Trusted Learner: An Improved Algorithm for Trusted Incremental Function Approximation. In: Proc. Symp. on Computational Intelligence in Dynamic and Uncertain Environments, pp. 16–24. IEEE Press, Paris (2011)Google Scholar
  3. 3.
    Dallaire, P., Besse, C., Chaib-draa, B.: Learning Gaussian Process Models from Uncertain Data. In: Leung, C.S., Lee, M., Chan, J.H. (eds.) ICONIP 2009, Part I. LNCS, vol. 5863, pp. 433–440. Springer, Heidelberg (2009)CrossRefGoogle Scholar
  4. 4.
    Denoeux, T.: Function Approximation in the Framework of Evidence Theory: A Connectionist Approach. In: Proc. Int. Conf. on Neural Networks, pp. 199–203. IEEE Press, Houston (1997)Google Scholar
  5. 5.
    Farrell, J., Polycarpou, M.: Adaptive Approximation Based Control. Wiley-Blackwell, Hoboken (2006)CrossRefGoogle Scholar
  6. 6.
    Girard, A., Murray-Smith, R.: Learning a Gaussian Process Model with Uncertain Inputs. Technical Report TR-2003-144, University of Glasgow (2003)Google Scholar
  7. 7.
    Lee, K.H.: Fuzzy Function. In: First Course on Fuzzy Theory and Applications. Advances in Soft Computing, vol. 27, pp. 153–170. Springer, Berlin (2005)CrossRefGoogle Scholar
  8. 8.
    Leonard, J.A., Kramer, M.A., Ungar, L.H.: Using Radial Basis Functions to Approximate a Function and its Error Bounds. IEEE Transactions on Neural Networks 3(4), 624–627 (1992)CrossRefGoogle Scholar
  9. 9.
    Lughofer, E., Guardiola, C.: Applying Evolving Fuzzy Models with Adaptive Local Error Bars to On-Line Fault Detection. In: Proc. of Genetic and Evolving Fuzzy Systems, Witten-Bommerholz, Germany, pp. 35–40. IEEE Press (2008)Google Scholar
  10. 10.
    Petit-Renaud, S., Denoeux, T.: Nonparametric Regression Analysis of Uncertain and Imprecise Data Using Belief Functions. Int. J. of Approximate Reasoning 35, 1–28 (2004)MathSciNetzbMATHCrossRefGoogle Scholar
  11. 11.
    Rasmussen, C.E., Williams, C.K.I.: Gaussian Processes for Machine Learning. MIT Press, Cambridge (2006)zbMATHGoogle Scholar
  12. 12.
    Tresp, V., Ahmad, S., Neuneier, R.: Training Neural Networks with Deficient Data. In: Advances in Neural Information Processing Systems, pp. 128–135. MIT Press, Denver (1993)Google Scholar
  13. 13.
    Tresp, V., Neuneier, R., Ahmad, S.: Efficient Methods for Dealing With Missing Data in Supervised Learning. In: Advances in Neural Information Processing Systems, pp. 689–696. MIT Press, Denver (1995)Google Scholar
  14. 14.
    Wang, L.X.: A Course in Fuzzy Systems and Control. Prentice-Hall, Inc., Upper Saddle River (1996)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Andreas Buschermöhle
    • 1
  • Jan Schoenke
    • 1
  • Werner Brockmann
    • 1
  1. 1.Smart Embedded Systems GroupUniversity of OsnabrückOsnabrückGermany

Personalised recommendations