HGM: A Novel Monte-Carlo Simulations based Model for Malware Detection

Abstract

Malware detection is a challenging and non-trivial task due to ever increase in several attacks and their sophistication level. Detection of such attacks demands the exploration of new approaches to generalize the attack patterns. One such approach is the use of Monte-Carlo simulations to train a reinforcement learning model. In this paper, we propose a self-adaptive Monte-Carlo simulation-based reinforcement model called Heuristic-based Generative Model (HGM), which generalizes the attack patterns in such a way that the new unknown attacks can be detected and flagged in real-time. The results show that HGM can detect a variety of malware with high accuracy.