Research Associate in Machine Learning for Complex Epidemics

Job summary
An exciting opportunity to join the EpiEng (Epidemiological Engineering) group led by Dr Kris Parag. We are a multidisciplinary group aiming to discover how principles from data science, mathematics and engineering can inform more realistic epidemic models and improve pandemic prediction and control algorithms at scale. We attempt to probe the limits of what models can achieve and optimise interventions by better understanding the feedback loops and network structures driving transmission.

As a Research Associate in Machine Learning for Complex Epidemics you will devise theory, models and methodology for tracking, forecasting and suppressing infectious disease outbreaks by creatively adapting algorithms and tools from reinforcement learning, multilayer network theory, causal statistics and complex systems. You will also develop open-access software packages and empirically validate outputs on real, heterogeneous datasets (spatiotemporal case and sequence data).

You will help answer questions about priority diseases (e.g., COVID-19, Ebola virus) and improve preparedness for an unknown Disease X by using machine learning and control theory to: better model poorly understood behavioural drivers of spread, extract salient features of transmission across scales (from individual to population levels) and optimise real-time decision-making from noisy data.

This role is collaborative and ideal for those with strong quantitative skills interested in confronting pandemic modelling challenges. You will engage with top academics at public health, engineering and data science departments at Imperial College and the Universities of Oxford, Cambridge, Hong Kong, Copenhagen and Lund. Experience in epidemiology is not essential. Candidates with varied backgrounds (e.g., physics, artificial intelligence, statistics, engineering or mathematics) are encouraged to apply.

Duties and responsibilities
You will:
-develop and analyse novel epidemic/pandemic models embedding key network and feedback architectures (validating these on real data)
-incorporate realistic behavioural, policy and other dynamics within these models across sociodemographic and spatial scales
-devise algorithms using reinforcement learning, feedback control and complex systems theory to optimise interventions and decision-making from noisy data
translate outputs into open-access R (or MATLAB/Python) packages that advance real-time pandemic control and prediction
-engage with multidisciplinary collaborators to improve model realism, practicality and flexibility (may require short research visits)
-present findings at conferences and meetings (locally and globally)
-publish research papers in leading, peer-reviewed journals

Essential requirements
You should possess:
-a PhD in a quantitative or computational discipline (e.g., machine learning/AI, statistics, mathematics, engineering, computer science, epidemiology)
-experience developing and implementing network and dynamical models
-knowledge of feedback control, network science and data science
-expertise fitting and validating stochastic models on spatiotemporal data
-experience coding and developing software
-a strong publication record in quantitative, refereed journals or conferences

Further information
The role is full-time and fixed term for 2 years, starting spring or later. Please apply online, including a CV with publications and statement (1-2 pages) on how your skills, and outputs make you suitable.

Candidates who have not yet been officially awarded their PhD will be appointed as a Research Assistant within the salary range £40,694 - £43,888 per annum.

Flexible working arrangements will be considered. For any queries, please contact k.parag@imperial.ac.uk.
For informal discussion about EpiEng work culture, email Dr Sandor Beregi s.beregi@imperial.ac.uk.

Hybrid working may be considered for this role. Staff working in roles that are suitable for hybrid working will normally be expected to work 60% of their time onsite. The opportunity for hybrid working will be discussed at interview.

For more information including how to apply please see https://www.imperial.ac.uk/jobs/search-jobs/description/index.php?jobId=18398&jobTitle=Research+Associate+