The paper is dedicated to the basic architecture of the artificial neural network of a new type – the Progressive Artificial Neural Network (PANN) – and its new training algorithm. The PANN architecture and its algorithm when applied together provide a significantly higher training speed than the known types of artificial neural networks and methods of their training.

In this paper, a new model of formal neuron, analog mechanisms of neuron training, and a new model of biological feedback are proposed. The statement is supported by the neurobiological data published by other authors and through our experiments in silicon.

Proposed are the new types of fast training, scalable analog and digital artificial neural networks (p-networks) based on the new model of formal neuron. P-network training time is dozens time faster than training time of the known networks.