Umjetne neuronske mreže
Artificial neural networks

Summary/Abstract: The paper presents the development of artificial neural networks and their importance for the development of artificial intelligence. It is anticipated that machines will reach human intelligence by the 2040s. Predictions are based on the theory of exponential increase of estimating capabilities of machines. The paper also presents the motif of updating the ideas of artificial neural networks, the difference between biological and artificial neural networks as well as inherent differences between the digital computer and the human brain. It presents the structure and functioning of biological and artificial neurons, as well as the characteristics of neural networks, in opposition to the conventional methods of data processing. Computers aren't able to stimulate the more complex structures of the human brain and therefore the conventional machine only mimics the neural network as a highly parallel architecture, while implementing conventional neural networks with conventional computers. The author opposed this possibility to virtual intelligence, machines that think instead of humans.

• Details
• Contents