Upcoming Seminar - SMC-Stan: A Scalable and Flexible Software tool for Better Bayesian Inference

Published on

Image Pixabay

The next upcoming seminar will be given by Professor Simon Maskell from the University of Liverpool who will talk about SMC-Stan, a software tool for better Bayesian Interference. Professor Maskell and his research was suggested by one of our LIV.DAT supervisors and we are very pleased that he has accepted our invitation to present.

About the talk

We are often interested in a setting where we can define a model for some data (i.e. a likelihood) and explicitly describe our prior knowledge such that we can pose the generic statistical inference task of inferring some parameters as an equivalent Bayesian inference problem. High-performance algorithms exist to make inferences from data using such models. These algorithms are complicated. It is therefore common to use probabilistic programming languages (PPLs) as a flexible mechanism for describing the models and then use pre-existing implementations of algorithms associated with those PPLs to make the inferences given the definition of the models (and some data). The key advantage of using a PPL with state-of-the-art algorithms is that you basically only need to define the model. This makes it possible for users of PPLs to be very much more ambitious in terms of the scale and complexity of the model that is considered. Work has now been done to generate next-generation algorithms and to integrate those algorithms into one PPL, Stan. These algorithmic enhancements focus on integration of Sequential Monte-Carlo (SMC) samplers, a new numerical Bayesian approach that offers substantial (i.e. >>100x) improvements in terms of run-time and therefore substantial potential to be ambitious with respect to the models considered.

In this talk, Prof Maskell will motivate the use of PPLs in general, explain a little about their focus on Stan specifically and describe why he thinks the advantages they are able to offer to users of SMC-Stan (now) have motivated significant interest and uptake from IBM, Google, GCHQ, epidemiologists working on COVID and UK government.



Simon Maskell is a Professor of Autonomous Systems at the University of Liverpool within the School of Electrical Engineering, Electronics and Computer Science. He completed his PhD at the Signal Processing Group of Cambridge University Engineering Department on ‘Sequentially Structured Bayesian Solutions’. Currently his research is focused on developing ground-breaking algorithmic solutions that can be translated into tangible advantage across multiple sectors in both industry and government. The solutions that Simon has developed have, for example, benefited organisations operating within defence, security, insurance, petrochemicals, pharma and transport.

Participation is free, but you need to register to attend this and other webinars in the series.

For more information and how to register please follow this link.


Upcoming Seminars


Professor Simon Maskell

Dept. of Electrical Engineering and Electronics, University of Liverpool

Seminar Title: “SMC-Stan: A Scalable and Flexible Software tool for Better Bayesian Inference”

Tuesday 23 February 2021 at 13:00 (Europe/London)



Seminar Title: “Applying data science methods to modernise transport and electricity infrastructures

March 2021 


Dr Anne O’Carroll

Remote Sensing Scientist, EUMETSTAT, Darmstadt (DE)

Seminar Title: “Combining satellite data with ocean surface measurements: Sea Surface Temperature (SST) observations”

Tuesday 20 April 2021 at 13:00 (Europe/London)


Professor Shirley Ho

Cosmology X Data Science Group, Flatiron Institute, New York (USA) &

Department of Astrophysical Sciences, Princeton University (USA)

Seminar Title: “Machine learning the Universe: Opening the Pandora Box”

Tuesday 25 May 2021 at 15:00 (Europe/London)