DIKUL - logo
Fakulteta za strojništvo, Ljubljana (FSLJ)
  • Adaptation of NN complexity to empirical information
    Grabec, Igor
    Instrumental observation of a natural phenomenon represents a transmission of information which is generally subject to random disturbances. In the article the scattering of empirical data provided ... by observation of a certain state is described by a probability distribution function. It is further applied at the estimation of probability distribution of a compound phenomenon which must be characterized by several possible states. The uncertainty of observation is commonly described by the information entropy. It is shown that the empirical information, which is defined as the difference between the entropies of a compound phenomenon and a single state, characterizes a complexity of observed phenomenon. With an increasing number of observations the value of increases less quickly as log and converges to a fixed value. A proper number of empirical samples sufficient to represent the phenomenon can be estimated as exp. The redundancy of empirical observations is therefore defined by the excess complexity log. When the phenomenon is modeled by a radial basis function NN a proper number of neurons can be described by the parameter Kr. An optimal NN structure can be obtained by minimizing the objective function which is comprised of the redundancy of the model and the information divergence between the representative and empirical distribution.
    Vir: Artificial neural nets and genetic algorithms : proceedings of the International Conference in Portorož, Slovenia, [April 6-9], 1999 (Str. 155-160)
    Vrsta gradiva - prispevek na konferenci
    Leto - 1999
    Jezik - angleški
    COBISS.SI-ID - 3256603

Vnos na polico