PhD studentships in Statistical Cyber Security
---------------------------------------------------

Please see below details of two open PhD studentships in the Statistics section of the Department of Mathematics, Imperial College London, in the area of statistical cyber-security. Details of how to apply can be found at the end.

---

PhD/Doctoral Studentship
Title: Data Enrichment and Anomaly Detection
Supervisors: Dr Nick Heard and Prof Niall Adams
Start Date: October 2017
End Date: April 2021

A 3.5 years EPSRC industrial CASE award PhD studentship in statistical cyber-security, co-funded by Leonardo (http://www.leonardocompany.com/en), is available within the Statistics Section at Imperial College London.

Data science techniques have the potential to provide the next generation of cyber-security defences. Inside a typical enterprise computer network, a number of high-volume data sources are available which could enable the discovery and prevention of cyber-attacks and other nefarious network activity. This project will seek to combine statistical analysis of computer network traffic flow data and host-based event logs with physical data such as access-control and other role-based indicators to build up an enriched picture.

The studentship comes with an enhanced stipend of over £21k per year, plus a generous research training support grant for equipment and attending international conferences. To be eligible for the full award, the applicant must either be a British citizen or an EU citizen who has resided in the UK for the last three years. See https://www.epsrc.ac.uk/skills/students/coll/icase/ for the full list of eligibility criteria. The student will spend one month each year at the offices of Leonardo, working with domain experts and obtaining hands-on experience with real data.

---

Title: Diagnostics for streaming statistical analytics
Supervisors: Prof Niall Adams and Dr Nick Heard
Start Date: October 2017
End Date: April 2021
 
A 3.5 years EPSRC/GCHQ funded PhD studentship in statistical cyber-security is available within the Statistics Section at Imperial College London.
 
Modern enterprise cyber-security systems often handle “streaming data” – an unending sequence of data reflecting aspects of the status of the enterprise network. In this context, statistical methods need to handle several problems: compute and memory efficient updating, and automatic adaptation to temporal variation. In such systems, the output from one statistical procedure may be the input to another. Naturally, this raises questions about the reliability of the output of statistical methods, and their role in downstream analysis. This project will develop novel methods of assessing and scoring the adequacy of streaming statistical procedures. Such scores provide the basis for both reconstructing the model and moderating the response to the output.
 
The studentship comes with a generous stipend, and support for travel and equipment. To be eligible for this award, the applicant must be a British citizen willing to acquire a level of security clearance. The student will spend one month each year working on classified problems, with domain experts to obtain hands-on experience with real data.

---

Applicants for either position are expected to have a masters level degree in a mathematical discipline containing a significant amount of statistics. The ideal candidate would be familiar with advanced statistical methods, possess strong computer programming skills, and have excellent interpersonal and communication skills.

To apply, candidates should complete an online application on the College website, 
http://www.imperial.ac.uk/study/pg/apply/how-to-apply/

Any queries should be directed to Dr Heard ([log in to unmask]) or Prof Adams ([log in to unmask]).

You may leave the list at any time by sending the command

SIGNOFF allstat

to [log in to unmask], leaving the subject line blank.