Application of reinforement machine learning in penetration testing

Myasnikov Alexey Vladimirovich

The article discusses the issues of applying reinforcement machine learning to the problem of penetration testing. Reinforcement machine learning algorithms require a specific representation of the environment in which they operate. The article describes an approach to representing the penetration testing process in terms of a Markov decision-making process, and also proposes an approach to finding the optimal attack path in the considered model using machine learning methods.

Keywords:machine learning, reinforcement learning, penetration testing, modeling of the penetration testing process, Markov decision making process.