Research Article
Deep CounterStrike: Counter Adversarial Deep Reinforcement Learning for Defense Against Metamorphic Ransomware Swarm Attack
@INPROCEEDINGS{10.1007/978-3-031-40467-2_3, author={Mohit Sewak and Sanjay K. Sahay and Hemant Rathore}, title={Deep CounterStrike: Counter Adversarial Deep Reinforcement Learning for Defense Against Metamorphic Ransomware Swarm Attack}, proceedings={Broadband Communications, Networks, and Systems. 13th EAI International Conference, BROADNETS 2022, Virtual Event, March 12-13, 2023 Proceedings}, proceedings_a={BROADNETS}, year={2023}, month={7}, keywords={Deep Reinforcement Learning Adversarial Learning Ransomware Metamorphic Malware Swarm Attack}, doi={10.1007/978-3-031-40467-2_3} }- Mohit Sewak
Sanjay K. Sahay
Hemant Rathore
Year: 2023
Deep CounterStrike: Counter Adversarial Deep Reinforcement Learning for Defense Against Metamorphic Ransomware Swarm Attack
BROADNETS
Springer
DOI: 10.1007/978-3-031-40467-2_3
Abstract
Ransomware, create a devastating impact when it infects a system. Fortunately, post the initial breach, such ransomware could be detected using advanced machine learning techniques, and therefore other high-value assets/systems can be protected from any repeat attack by the same ransomware. However, using metamorphism, advanced/ second-generation ransomware can alter its structure after every successful infection. With this ability of metamorphism, such advanced ransomware could continue to evade any defensive mechanism and keep infecting systems in subsequent networks. Currently, there exists neither any proven defensive mechanism nor any useful dataset to train a defensive mechanism against such advanced ransomware. Therefore, we present a deep counter adversarial reinforcement learning-based system that learns how to normalize the metamorphism of such advanced ransomware to design a credible defence against such advanced attacks. To augment training data for this system, we design and develop a deep adversarial reinforcement learning solution, to generate swarms of such advanced ransomware.
