Torben Sell

Position: Postdoctoral Research Assistant
Field of research: High-dimensional computational statistics
E-mail: torben.sell@ed.ac.uk

School of Mathematics,
University of Edinburgh,
James Clerk Maxwell Building,
Peter Guthrie Tait Road,
Edinburgh, EH9 3FD,
United Kingdom

---

About

I am a postdoctoral research assistant at the University of Edinburgh working with Dr Tim Cannings.

Before joining the School of Mathematics in Edinburgh, I was a PhD student at the Cantab Capital Institute for the Mathematics of Information (CCIMI) and the Statistical Laboratory within the Department of Pure Mathematics and Mathematical Statistics (DPMMS) at the University of Cambridge, supervised by Dr Singh.

In April 2021, I moved to the University of Edinburgh with a LMS Early Career Fellowship, and started my current position in May 2021.

---

Research

My research interests include

• Statistical methodology
• Bayesian statistics
• High-dimensional statistics
• Statistical learning
• Markov chain Monte Carlo
• Piecewise deterministic Markov processes
• Bayesian neural networks
• Particle filtering
• Stochastic control
• Function space sampling methods

I gave a short (5 min) introduction to particle filtering at the Isaac Newton Institute in May 2018, which is available online via this link, and a slightly longer talk on another project in November 2018, found here. A high-level, and hopefully entertaining, explanation of one of my research projects is available on YouTube.

---

Teaching

I am currently tutoring and lecturing part of the Facets of Mathematics course at the University of Edinburgh.
Previously I taught the following courses:

• 2017-2020 - Supervisor for Part II Topics in Analysis and Part II Numerical Analysis at the University of Cambridge
• 2014-2016 - Tutor for Analysis I and II at the University of Rostock

---

Publications

1. TS, "Advanced Bayesian Monte Carlo Methods for Inference and Control", PhD Thesis, pdf available upon request


2. TS, Sumeetpal Sidhu Singh, "Dimension-Robust Function Space MCMC With Neural Network Priors", under review, arXiv preprint available


3. Jacob Vorstrup Goldman*, TS*, Sumeetpal Sidhu Singh, "Gradient-Based Markov Chain Monte Carlo for Bayesian Inference With Non-Differentiable Priors", accepted for publication by the Journal of the American Statistical Association, arXiv preprint available. * denotes equal contributions


4. TS, "MCMC Methods for Functions: Modifying Old Algorithms to Make Them Faster", Essay for Part III of the Mathematical Tripos, pdf


5. TS, "A Convergence Analysis for Preconditioned Gradient Type Eigensolvers", Thesis for Bachelor of Science, pdf

---

Academic Activities

In 2020 I co-organised the Mathematics of Data Science conference, and wrote a short lessons learned paragraph about it.

In the academic year 2018/2019 I was one of two directors of the Part III seminar series at the University of Cambridge.

I enjoy reviewing papers within my area of expertise.