Wide learning: Using an ensemble of biologically-plausible spiking neural networks for unsupervised parallel classification of spatio-temporal patterns

Abstract

Spiking neural networks have been previously used to perform tasks such as object recognition without supervision. One of the concerns relating to the spiking neural networks is their speed of operation and the number of iterations necessary to train and use the network. Here, we propose a biologically plausible model of a spiking neural network which is used in multiple, separately trained copies to process subsets of data in parallel. This ensemble of networks is tested by applying it to the task of unsupervised classification of spatio-temporal patterns. Results show that despite different starting weights and independent training, the networks produce highly similar spiking patterns in response to the same class of inputs, enabling classification with fast training time.