10 October 2022,
by Rob Flood, University of Edinburgh
Hi there! I’m Rob, a Ph.D student studying at the University of Edinburgh. As part of my Saltire project, I spent one month in Budapest, Hungary.
My research is primarily concerned with synthetic data generation for use in security related analysis. By its very nature, real-world security data is sensitive. Legislative frameworks such as GDPR or the DPA limit researchers’ abilities to capture and release such data, for good reason. It may contain personally-identifiable information or provide malicious actors with important intelligence about an organisation’s security capabilities. As a result, researchers often rely on synthetic datasets. However, despite the explosion of research into machine learning applied to security, there has been little work discussing how these datasets should be designed, generated and released. In contrast, my work argues that, because malicious data, network topologies and threat models are all highly variable, great care needs to be taken when building these datasets.
I undertook my visit with Prof. Levente Buttyan at the Crysys Lab at the Budapest University of Technology and Economics. During this time, we focused on the generation of synthetic data for Industrial Control Systems, using data from the lab’s data generation testbed. Specifically, we discussed ways in which we could adapt domain randomisation, a technique common in other areas of synthetic data generation, to the field of network security.
I’ve never visited Hungary and had little idea of what to expect. My trip introduced me to Hungarian cuisine, architecture and history, all of which I enjoyed thoroughly. Moreover, my host treated me graciously, showing me his hometown of Szentendre and showing me where to get the best Lángos in Hungary.
Despite the visit lasting only a month, the scope of this work has expanded considerably, bringing up a number of new research questions to explore in the coming months. This SICSA exchange was extremely fruitful for me and I’m very grateful for being given the opportunity.