Summary of Multi-agent Reinforcement Learning For Maritime Operational Technology Cyber Security, by Alec Wilson et al.
Multi-Agent Reinforcement Learning for Maritime Operational Technology Cyber Security
by Alec Wilson, Ryan Menzies, Neela Morarji, David Foster, Marco Casassa Mont, Esin Turkbeyler, Lisa Gralewski
First submitted to arxiv on: 18 Jan 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Cryptography and Security (cs.CR); Multiagent Systems (cs.MA)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This research paper introduces a simulation environment for industrial control systems and explores the application of Multi-Agent Reinforcement Learning (MARL) to autonomous cyber defence decision-making on maritime-based Integrated Platform Management Systems (IPMS). The study demonstrates the potential for MARL to outperform traditional IT-centric cyber defence solutions, highlighting the importance of developing robust and adaptable defences for Operational Technology (OT) infrastructure. The paper presents experimental results showing that a shared critic implementation of MAPPO (Multi Agent Proximal Policy Optimisation) outperformed Independent Proximal Policy Optimisation (IPPO), achieving an optimal policy after 800K timesteps. Hyperparameter tuning also significantly improved training performance. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This research shows how to use artificial intelligence to help protect industrial control systems from cyber attacks. The team created a simulated environment for these systems and tested different ways of using machine learning algorithms to make decisions about when to take action against potential threats. They found that one approach, called MAPPO, was better than another at making good choices most of the time. This is important because industrial control systems are often more vulnerable to cyber attacks than other types of computer systems. |
Keywords
* Artificial intelligence * Hyperparameter * Machine learning * Reinforcement learning
