Description

Register for this Event

This work fits into the context of digital twins, which are usually made of two components: a model and some data. When developing a digital twin, many fundamental questions exist, some connected with the data and its reliability and uncertainty, and some to do with dynamic model updating. To combine model and data, we use Data Assimilation (DA). DA is the approximation of the true state of some physical system by combining real-world observations with a dynamic model. DA models have increased in sophistication to better fit application requirements and circumvent implementation issues. Nevertheless, these approaches are incapable of fully overcoming their unrealistic assumptions. Machine Learning (ML) shows great capability in approximating nonlinear systems and extracting meaningful features from high-dimensional data. ML algorithms can assist or replace traditional forecasting methods. However, the data used during training in any ML algorithm include numerical, approximation and round off errors, which are trained into the forecasting model. Integration of ML with DA increases the reliability of prediction by including information in real time and with a physical meaning. This talk introduces Data Learning, a field that integrates Data Assimilation and Machine Learning to overcome limitations in applying these fields to real-world data. We present several Data Learning methods and results for some test cases, though the equations are general and can easily be applied elsewhere.


About the speaker

Dr Rossella Arcucci has been with the Data Science Institute at Imperial College since 2017, where she has created the Data Assimilation and Machine Learning (DataLearning) Working Group. The group is now a focal point for researchers and students of several departments at Imperial and other Universities in UK and Europe. She leads and coordinate the group and she supervises students, PhD students and Post-Doc Researchers.

She collaborates with the Leonardo Centre at Imperial College Business School, where she contributes to the development of integrative, just and sustainable models of economic and social development by discovering, testing and diffusing new logics of business enterprise.

The models Rossella has developed have produced impact in many applications such as finance (to estimate optimal parameters of economic models), social science (to merge twitter and pooling data to better estimate the sentiment of people), engineering (to optimise the placement of sensors and reduce the costs), geoscience (to improve accuracy of forecasting), climate changes and others. She has developed accurate and efficient models with data analysis, fusion and data assimilation for incomplete, noisy or Big Data problems, always including uncertainty quantifications and minimizations.

She works on numerical and parallel techniques for accurate and efficient Data Assimilation and Machine Learning models. Efficiency is achieved by virtue of designing models specifically to take full advantage of massively parallel computers.

She finished her PhD in Computational and Computer Science in February 2012. She received the acknowledgement of Marie Skłodowska-Curie fellow from the European Commission Research Executive Agency in Brussels in February 2017.

Registration

This event will be a HYBRID event. To join this seminar online OR in person, please register on Zoom. You will then receive an email with the dial in details for those attending virtually. Please check your spam/junk folders for joining details.

Registration: https://us02web.zoom.us/meetin...

Contact complexity@inet.ox.ac.uk for more information.

Venue

Research Programmes