Features of building neural networks taking into account the specifics of their training to solve the tasks of searching for network attacks

Abstract: The problems of building neural networks to solve the prob-lems of detecting network intrusions, taking into account mod-ern publicly available technologies, are considered. Several configurations of neural networks are analyzed: a simple per-ceptron, a combined network consisting of two interconnected networks, simplified networks based on a simple perceptron, LSTM networks using hidden layers with data compression function. The weaknesses and strengths of neural network ar-chitectures are considered, taking into account the specifics of their training based on abnormal traffic datasets in intrusion detection tasks.

Keywords: network attack, neural network, dataset, feature matrix, activa-tion function