In a radial basis function (RBF) network, the RBF centers and widths can be evolved by a cooperative-competitive genetic algorithm. The set of genetic strings in one generation of the algorithm represents one REP network, not a population of competing networks. This leads to moderate computation times for the algorithm as a whole. Selection operates on individual RBFs rather than on whole networks. Selection therefore requires a genetic fitness function that promotes competition among RBFs which are doing nearly the same job while at the same time promoting cooperation among RBFs which cover different parts of the domain of the function to be approximated. Niche creation resulting from a fitness function of the form |w(i)|(beta)/E(|w(i')|(beta)), 1

An evolutionary neural network training algorithm is proposed for radial basis function (RBF) networks. The locations of basis function centers are not directly encoded in a genetic string, but are governed by space-filling curves whose parameters evolve genetically. This encoding causes each group of codetermined basis functions to evolve to fit a region of the input space. A network produced from this encoding is evaluated by training its output connections only. Networks produced by this evolutionary algorithm appear to have better generalization performance on the Mackey-Glass time series than corresponding networks whose centers are determined by k-means clustering.

Purely parallel neural networks can model object recognition in brief displays – the same conditions under which illusory conjunctions (the incorrect combination of features into perceived objects in a stimulus array) have been demonstrated empirically (Treisman 1986; Treisman & Gelade 1980). Correcting errors of illusory conjunction is the “tag-assignment” problem for a purely parallel processor: the problem of assigning a spatial tag to nonspatial features, feature combinations, and objects. This problem must be solved to model human object recognition over a longer time scale. Our model simulates both the parallel processes that may underlie illusory conjunctions and the serial processes that may solve the tag-assignment problem in normal perception. One component of the model extracts pooled features and another provides attentional tags that correct illusory conjunctions. Our approach addresses two questions: (i) How can objects be identified from simultaneously attended features in a parallel, distributed representation? (ii) How can the spatial selectional requirements of such an attentional process be met by a separation of pathways for spatial and nonspatial processing? Our analysis of these questions yields a neurally plausible simulation of tag assignment based on synchronizing feature processing activity in a spatial focus of attention.

No abstract

A well-performing set of radial basis functions (RBFs) can emerge from genetic competition among individual RBFs. Genetic selection of the individual RBFs is based on credit sharing which localizes competition within orthogonal niches. These orthogonal niches are derived using singular value decomposition and are used to apportion credit for the overall performance of the RBF network among individual nonorthogonal RBFs. Niche-based credit apportionment facilitates competition to fill each niche and hence to cover the training data. The resulting genetic algorithm yields RBF networks with better prediction performance on the Mackey-Glass chaotic time series than RBF networks produced by the orthogonal least squares method and by k-means clustering.