A predictive modular neural network method is applied to the problem of unsupervised time-series segmentation. The method consists of the concurrent application of two algorithms: one for source identification, the other for time-series classification. The source identification algorithm discovers the sources generating the time series, assigns data to each source, and trains one predictor for each source. The classification algorithm recursively computes a credit function for each source, based on the competition of the respective predictors, according to their predictive accuracy; the credit function is used for classification of the time-series observation at each time step. The method is tested by numerical experiments.

An incremental credit assignment (ICRA) scheme is introduced and applied to time series classification. It has been inspired from Bayes' rule, but the Bayesian connection is not necessary either for its development or proof of its convergence properties. The ICRA scheme is implemented by a recurrent, hierarchical, modular neural network, which consists of a bank of predictive modules at the lower level, and a decision module at the higher level. For each predictive module, a credit function is computed; the module that best predicts the observed time series behavior receives highest credit. We prove that the credit functions converge (with probability one) to correct values. Simulation results are also presented.