Alexey Redozubov

Vasily Morzhakov, Alexey Redozubov

Cortical minicolumns are considered a model of cortical organization. Their function is still a source of research and not reflected properly in modern architecture of nets in algorithms of Artificial Intelligence. We assume its function and describe it in this article. Furthermore, we show how this proposal allows to construct a new architecture, that is not based on convolutional neural networks, test it on MNIST data and receive close to Convolutional Neural Network accuracy. We also show that the proposed architecture possesses an ability to train on a small quantity of samples. To achieve these results, we enable the minicolumns to remember context transformations.

Alexey Redozubov

The structure of the axon-dendrite connections of neurons of the brain creates a rich spatial structure in which provided various combinations of signals surrounding neurons. Structure of dendritic trees and shape of dendritic spines allow repeatedly increase combinatorial component through cross synapses influence neighboring neurons. In this paper it is shown that the diffuse spreading of neurotransmitters allows neurons to detect and remember significant set of environmental activity patterns. As a core element fixation described extrasynaptic metabotropic receptive clusters. The described mechanism leads to the appearance of wave processes, based on the propagation of the front-line areas of spontaneous activity. In the proposed model, any compact pattern of neural activity is seen as a source emitting a diverging wave endogenous spikes. It is shown that the spike pattern of the wave front is strictly unique and uniquely defined pattern that started the wave. The propagation of waves with a unique pattern allows anywhere in nature undergoing brain wave patterns there to judge the whole brain processes information. In these assumptions naturally described mechanism of projection information between regions of the cortex. Performed computer simulations show the high effectiveness of such information model.