Abstract
The possibility of using artificial neural networks to search for a rare \(pp \to tHbq\) process at the Large Hadron Collider is considered. An evolutionary algorithm that makes it possible to construct a neural network with an optimal set of parameters has been developed. Using Monte Carlo simulation, the significance of signal separation from the background is assessed. The use of neural networks significantly increases the signal significance and makes it possible to detect the \(pp \to tHbq\) process based on the data from completed sessions of the Large Hadron Collider.
