Gamma-filter self-organising neural networks for unsupervised sequence processing

Abstract

Adding gamma-filters to self-organising neural networks for unsupervised sequence processing is proposed. The proposed gamma-context model is applied to self-organising maps and neural gas networks. The gamma-context model is a generalisation that includes as a particular example the previously published merge-context model. The results show that the gamma-context model outperforms the merge-context model in terms of temporal quantisation error and state-space representation.