I'm very much looking forward to speaking at the upcoming workshop on MCMC at Newcastle. Details here: sites.google.com/vie... I will speak about MCMC with active subspaces arxiv.org/abs/2501.0...

I will be a supervisor for an upcoming PhD! (with Kirsty Hassall and Jacqueline Stroud) The opportunity is an EPSRC DLA Interdisciplinary Scholarship: "Unravelling the sound underground: Detecting and estimating earthworm abundance using low-frequency vibrations".
warwick.ac.uk/fac/sc...

Two interesting upcoming events: Data-Centric Engineering at Cambridge on 15th Jan sites.google.com/vie... and the 1st International Workshop on Bayesian Approach in Civil Engineering in Hong Kong events.polyu.edu.hk/...

At the University of Birmingham today, speaking to CENTA PhD students about "Understanding Machine Learning and Artificial Intelligence". https://centa.ac.uk/

Three of my PhD students have their viva in the next 18 days. "Go well" to all of them. I'll try to post a few things about their work in the next few weeks.

Since the ggsmc plotting functions are just wrappers about ggplot2, it is easy to change the style or add to the plot (here, a line showing the true target position). Here we also automatically generate a title for the figure.
3/3

This figure shows the output from a particle filter, filtering on a univariate state using a constant velocity model in a target tracking application.
2/3

More on ggsmc... The "genealogy" function visualises the particle representation of an SMC or ensemble Kalman algorithm over the iterations of the algorithm. This figure shows the output of an SMC sampler with annealing with a mixture target distribution.
1/3

... and a scatter plot.
3/4

then a density...
2/4

In this thread are some examples of figures generated using my new R package ggsmc. Each figure shows output from one target in a particle filter. To begin we have a histogram.
1/4