Recognition of spatio-temporal patterns by a multi-layered neural network model

Using a multilayered neural network model to recognize spatio-temporal patterns is proposed. The hierarchical network used in this model consists of three kinds of neuron-like cells: C-cells, which absorb positional errors, D-cells which allow for time distortions, and S-cells which extract specific spatio-temporal features. In the hierarchical network, local spatio-temporal features of the input pattern are extracted by cells of the lower stages. These are gradually integrated into more global features in the higher stages. During this process of extracting and integrating features, both positional errors and time distortions are gradually tolerated. Finally, each cell of the highest stage integrates all of the information of the spatio-temporal input pattern, and responds to only one specific pattern.