This short event series ran over three days and the aim was to run this as in-person events. There was  a virtual option for those unable to attend physically, but we welcomed many of you back to Cambridge.

Background

There is an urgent need to advance the reliability and reproducibility of all forms of computer-based calculations.  In an experimental setting, it is often common practice to provide not only measured values themselves, but also an estimate of the uncertainty in the measurements.  Yet this is not yet standard practice when presenting predictions made by computational models. 

Computational methods are becoming increasingly important for decision-making and thus a similar approach is needed to make computational predictions actionable, since virtually all models are contaminated by several sources of uncertainty.  Sources of uncertainty originate in the imperfectly known input parameters assumptions made within the mathematical form of the model, and the influence of stochastic effects. Uncertainty quantification is a challenging field that combines scientific models, statistics, high-performance computing, and increasingly machine learning methods.

The mathematical sciences have a key role to play in addressing these challenges. In this event series, the Newton Gateway worked with members of the EPSRC-funded SEAVEA project. The work of the project seeked to address such challenges through the promotion and development of an open-source software toolkit for verification, validation and uncertainty quantification (VVUQ). This is optimised for efficient execution on many platforms including those at the current petascale and the emerging exascale.  These capabilities offer new opportunities for simulations in fields as diverse as fusion, weather and climate modelling, epidemiology, advanced materials, biomedicine and many other domains.

Aims and Objectives

A series of three half-day science events were presented between 8^th -10^th February 2022 which highlighted key aspects of the research, brought together relevant stakeholders across academia, industry and government. It was delivered as a RAMP Continuity Network activity.

The event series focused on uncertainty modelling for epidemiology and pandemics, and highlighted new opportunities for simulations in fields as diverse as fusion, weather and climate modelling, advanced materials, biomedicine and many other domains. Sessions were spread across three half days as follows:

• VVUQ Fundamentals with a Focus on Epidemiology and Pandemics
• Wider Applications of VVUQ Methods
• Challenges and Advanced Methods for Epidemiological Modelling

There was the opportunity to join two one-hour tutorials on the open-source software toolkits for VVUQ. These ran in the mornings on days two and three.

You can view the Programme here.

Registration

Registration for this event is now closed.