Analysis of perspective models of artificial neural networks for control of robotic objects

Abstract

Artificial neural networks are used in various fields of science: from speech recognition systems to recognition of the secondary protein structure, classification of various types of cancer and genetic engineering. However, how do they work and what are they good at? When it comes to tasks other than processing large amounts of information, the human brain has a big advantage compared to a computer. A person can recognize faces even if there are many foreign objects and poor lighting in the room. We easily understand strangers even when we are in a noisy room. But, despite years of research, computers are still far from performing such tasks at a high level.