The knowledge gained from the pre-lab and hands-on lab in this module allows students to further explore the practical use of Malware Analysis, specifically the Deep Reinforcement Learning algorithm.
Anomaly-based IDS:
Recall in the pre-lab of this module, we analyzed some of the algorithms to implement malware analysis with Generative AI. They included Deep Reinforcement Learning, Long-Short Term Memory, and Recurrent Neural Networks.
Using your knowledge from the Malware Analysis Module, demonstrate Deep Reinforcement Learning (DRLs) in network traffic. This means that you must implement a learning algorithm and an exploration-exploitation strategy with Generative AI. The algorithms may be DQNs, PPO, or A3C. Refer to the references at the bottom of this page for more information about these algorithms. You may use Google Colab to implement this.
References:
Mnih, V., Kavukcuoglu, K., Silver, D., Graves, A., Antonoglou, I., Wierstra, D., & Riedmiller, M. "Playing Atari with deep reinforcement learning." arXiv, Cornell University. December 19, 2013. https://doi.org/10.48550/arXiv.1312.5602
Schulman, J., Wolski, F., Dhariwal, P., Radford, A., & Klimov, O. "Proximal policy optimization algorithms." arXiv, Cornell University. August 28, 2017. https://doi.org/10.48550/arXiv.1707.06347
Sciforce. "Reinforcement learning and asynchronous actor-critic agent (A3C) algorithm, explained." Medium. March 25, 2021. https://medium.com/sciforce/reinforcement-learning-and-asynchronous-actor-critic-agent-a3c-algorithm-explained-f0f3146a14abÂ