Choosing the Reinforcement Learning Method for Modeling DDos Attacks
P. Cheskidov, Kseniia Yu. Nikolskaia, Aleksey Minbaleev · 2019 International Multi-Conference on Industrial Engineering and Modern Technologies (FarEastCon) · 2019
Distributed denial of service (DDoS) attacks constitute a rapidly evolving threat in the current Internet. In the field of DDoS attacks, as in all other areas of cybersecurity, the battle between the shield and the sword does not stop. Attackers are using increasingly sophisticated methods. Solution providers follow them, releasing new products in order to prevent malicious intent. Old tools stop working, new approaches and tools are required in order not to become a victim of cybercriminals. This document discusses the development path that tools for countering DDoS attacks go through under the influence of changing cybercriminal approaches. This paper discusses reinforcement training methods, gives a brief description of them, and examines the scope of reinforcement training in defense against DDoS attacks. The terminology of reinforced learning is also described, and an agent-based learning algorithm is described.