Join us for this seminar by Mauricio Álvarez (Manchester) as part of the North West Seminar Series in Mathematical Biology and Data Sciences. More details about the joint series can be found here https://northwestseminars.great-site.net/ .
The talk will be hosted in person in room 3.40 of the Simon Building. For those who cannot attend in person the talk will also be streamed via zoom, please contact carl.whitfield@manchester.ac.uk or igor.chernyavsky@manchester.ac.uk for the zoom link, or sign up to the mailing list.
Title: Bridging the gap between data-driven and mechanistic modelling: the latent force model approach.
Abstract: Current machine learning models tend to be purely data-driven models where computation and data size are key requirements. By encoding physics into machine learning models, either in the form of ordinary or partial differential equations, we can develop data-efficient approaches to data modelling that in many cases can provide further insights into the data-generation mechanism. In this talk, I will introduce a family of such physics-inspired machine learning models, which we have coined as Latent Force Models. A latent force model is a Gaussian process with a covariance function inspired by a differential operator. Such a covariance function is obtained by performing convolution integrals between Green's functions associated with the differential operators and covariance functions associated with latent functions. Latent force models have been used in several fields for grey box modelling and Bayesian inversion. In this talk, I will introduce latent force models and several recent works in my group where we have extended this framework to non-linear problems.
Bio:
Mauricio A. Álvarez is a Senior Lecturer in Machine Learning in the Department of Computer Science at the University of Manchester. Before joining Manchester, he was an Associate Professor at Universidad Tecnológica de Pereira, Colombia, and a Senior Lecturer at the University of Sheffield. He is interested in machine learning in general, its interplay with mathematics and statistics and its applications. In particular, his research interests include probabilistic models, kernel methods, and stochastic processes. He works on the development of new probabilistic models and their application in different engineering and scientific areas including Neuroscience, Neural Engineering, Systems biology, and Humanoid Robotics, among others. He is internationally known for his work on multi-output Gaussian processes and physically inspired probabilistic modelling.
Dr Álvarez is the Director of a new AI Centre for Doctoral Training on Decision Making for Complex Systems in Manchester. He is Associate Editor for the Transactions on Machine Learning OpenReview journal. He has been area chair for several machine learning conferences, including the Advances in Neural Information Processing Systems (NeurIPS) conference, the Uncertainty in Artificial Intelligence (UAI) conference, the International Conference on Learning Representations (ICLR), the AAAI Association conference and the Artificial Intelligence and Statistics (AISTATS) conference. He has been the main organiser of the Gaussian process summer school in the UK since 2017 and co-organiser of the ELLIS (European Laboratory for Learning and Intelligent Systems) Summer School on Machine Learning for Healthcare and Biology.
To subscribe to the mailing list for this event series, please send an e-mail with the phrase “subscribe math-lifesci-seminar” in the message body to listserv@listserv.manchester.ac.uk