Abstract: In this talk, we will discuss the benefits of Bayesian methods for uncertainty modelling and quantification in challenging imaging/sensing inverse problems. In particular, we will discuss the problem of detecting and quantifying sources in linear mixtures, potentially corrupted by anomalies. While Markov chain Monte Carlo (MCMC) methods are preferred tools for uncertainty quantification in complex Bayesian models, their computational complexity generally becomes prohibitive for high-dimensional problems or model order selection problems requiring real-time analysis. Thus, in this talk we will also discuss more scalable, approximate methods based on expectation-propagation, enabling robust and faster inference for radiation source unmixing and providing better results than optimization methods based on convex relaxation.


Bio: Dr Altmann completed his Ph.D. within the Signal and Communication Group of the IRIT Laboratory in Toulouse, France, in 2013. In 2014, He was awarded a postdoctoral Fellowship by the Direction Générale de l’Armement and joined Heriot-Watt University to develop computational methods for hyperspectral imaging and Lidar-based ranging applications. Since February 2017, he has been an Assistant Professor at the School of Engineering and Physical Sciences, Heriot-Watt University. His research interests include Bayesian modelling and computation (Monte Carlo, variational inference) for imaging/sensing inverse problems, with applications in biomedical imaging, remote sensing and nuclear sciences.

If you have a question about this list, please contact:

Each talk has an organiser. Please contact them in the first instance if you have a query about a particular talk. Only contact one of the people below if you have a question about the list, such as whether your talk or series could be added.