Heterogeneous radial basis function networks

@article{Wilson1996HeterogeneousRB,
  title={Heterogeneous radial basis function networks},
  author={D. Randall Wilson and Tony R. Martinez},
  journal={Proceedings of International Conference on Neural Networks (ICNN'96)},
  year={1996},
  volume={2},
  pages={1263-1267 vol.2}
}
Radial basis function (RBF) networks typically use a distance function designed for numeric attributes, such as Euclidean or city-block distance. This paper presents a heterogeneous distance function which is appropriate for applications with symbolic attributes, numeric attributes, or both. Empirical results on 30 data sets indicate that the heterogeneous distance metric yields significantly improved generalization accuracy over Euclidean distance in most cases involving symbolic attributes. 

Figures and Topics from this paper.

Citations

Publications citing this paper.

References

Publications referenced by this paper.
SHOWING 1-10 OF 11 REFERENCES

An Empirical Comparison of Discretization Models,

Ventura, Dan, Tony Martinez
  • Proceedings of the 10th Pntemational Symposium on Computer and Information Sciences,
  • 1995

Aha , “ Towards a Better Understanding of Memory - Based and Bayesian Classifiers

John Rachlin, Simon Kasif, Steven Salzberg, W David
  • Proceedings of the Eleventh International Machine Learning Conference
  • 1994

An Overview of Evolutionary Computation,

Spears, M William, +3 authors Hugo de Garis
  • Proceedings of the European Conference on Machine Learning,
  • 1993

Enhancements to probabilistic neural networks

  • [Proceedings 1992] IJCNN International Joint Conference on Neural Networks
  • 1992

How Gaussian radial basis functions work

  • IJCNN-91-Seattle International Joint Conference on Neural Networks
  • 1991

Phoneme classification experiments using radial basis functions

  • International 1989 Joint Conference on Neural Networks
  • 1989