Self-organization and association for temporal coding

This article discusses the unsupervised learning of a network for a temporally precise sequence. A network of leaky neurons with many excitatory random inputs is able to learn a fine spatio-temporal pattern, by having the neurons select their connections. The trained network works as an associative memory or a filter in distinguishing a temporal sequence with high precision. Distinguishes the training sequence through filtering the disarranged sequence according to its correlation value from the training sequence