Artificial Astrocyte Networks, as Components in Artificial Neural Networks

摘要

Recent findings in neurophysiology provided evidence that not only neurons but also networks of glia-astrocytes are responsible for processing information in the human brain. Based on these new findings, information processing in the brain is defined as communication between neurons-neurons, neurons-astrocytes and astrocytes-astrocytes. Artificial neural networks (ANNs) model the neuron-neuron communications. Artificial neuron-glia networks (ANGN), in addition to neuron-neuron communications, model neuron-astrocyte connections. This research introduces a new model of ANGN that captures these three possible communications. In this model, random networks of artificial glia astrocytes are implemented on top of a typical neural network. The networks are tested on two classification problems, and the results show that on certain combinations of parameter values specifying astrocyte connections, the new networks outperform typical neural networks. This research opens a range of possibilities for future work on designing more powerful architectures of artificial neural networks that provide more realistic models of the human brain.