Mathematical Innovation Research Assistant (fixed-term)

Project Overview

This project is at the intersection of machine learning, differential equations, and numerical weather prediction (NWP). 

This collaboration with the Met Office integrates advanced mathematical models with state-of-the-art machine learning techniques to develop forecasting models that are more accurate, efficient, and capable of incorporating complex atmospheric phenomena.

This is a fixed-term position for 3 months. 

Background

The core of our project is inspired by recent advancements in computational and atmospheric sciences to enhance traditional forecasting models with innovative computational techniques. 

Our strategy involves reimagining the conventional architecture of weather prediction models by incorporating novel mathematical approaches that allow for a more nuanced representation of atmospheric processes. 

The intention is to create models that can more accurately predict weather patterns, incorporating a broad range of atmospheric dynamics and conservation laws.

Key Responsibilities

• Develop and refine machine learning models that effectively integrate differential equations with NWP for enhanced forecasting capabilities.
• Collaborate with the Met Office for access to and utilisation of weather data.
• Implement and evaluate the model using Python, ensuring compatibility with ML frameworks and Firedrake software.

Desired Qualifications

• Enrolment in or completion of a PhD in a relevant field, with strong competencies in machine learning, differential equations, and NWP.
• Proficiency in Python and familiarity with ML frameworks such as PyTorch.
• Interest or experience in atmospheric science, especially in applying mathematical models to weather forecasting challenges.
• Excellent collaborative and communication skills for effective teamwork and complex concept explanation.

 Additional Perks

• Direct collaboration with the Met Office, offering unique insights into the practical applications of forecasting technologies.
• Opportunity to make significant contributions to the advancement of weather prediction technology, benefiting both academic research and practical applications.
• A chance to bridge theoretical research with real-world implementation, enhancing the candidate's experience and professional development.

For informal enquires about this role, please contact the Prof Tristan Pryer, Director of IMI: [email protected] .