“Combining statistics with hard core engineering and applied mathematics is for me just absolutely fantastic”: An interview with Mark Girolami

Mark Girolami is an EPSRC Established Career Research Fellow (2012 – 2018) and previously an EPSRC Advanced Research Fellow (2007 – 2012). He is the Director of the £10M Lloyds Register Foundation – Turing Programme on Data Centric Engineering and previously led the EPSRC funded Research Network on Computational Statistics and Machine Learning. In 2011 he was elected to the Fellowship of the Royal Society of Edinburgh when he was also awarded a Royal Society Wolfson Research Merit Award. In 2018 he was awarded the Royal Academy of Engineering Research Chair on Data Centric Engineering.

He was one of the founding Executive Directors of the Alan Turing Institute for Data Science from 2015 to 2016 before taking leadership of the Data Centric Engineering Programme at The Alan Turing Institute, as well as Professor in the Department of Mathematics at Imperial College, London.

Alison Oliver talks to Professor Girolami about his career.

1. You studied Mechanical Engineering at University of Glasgow followed by a PhD at Paisley. What was it that first introduced you to mechanical engineering as a discipline?

I studied Mechanical Engineering because I really enjoyed engineering when I was at school, and it is as simple as that. I enjoyed maths and I enjoyed engineering, mechanisms and so forth, and there was no real rhyme or reason, no thought to it. I actually obtained the PhD 10 years after. Once I graduated, I spent 10 years with IBM, and left to do the PhD at Paisley because it was close to where I lived. That was it and I thought it was great.

2. You were then an engineer at IBM before taking up a career as a lecturer and later returning to Paisley as Professor of Computing Science, followed by Glasgow where you eventually became Chair in Computing Science. What inspired this move to lecturing?

I think to be honest, that was always my aspiration: to work as an academic. When I went to IBM, it was really with the intent of maybe getting a few years’ experience and those few years ended up being a decade! Probably my original plan was to spend two or three years to get some professional experience before getting a PhD, but the way things worked out, it took quite a lot longer before I actually did it. But the inspiration was always there, really.

3. In 2010, you were appointed Chair of Statistics at Department of Statistical Science at UCL? What led you to this move?

The Department of Statistical Science at UCL is one of the historic departments of statistics in the world. It was the first in the world, and it was founded by Karl Pearson; Fisher had been there; and then Pearson’s son was there. There’s this rich history, and when the opportunity arose for me to move to UCL to this really historic department, I thought I’ve got to go, I can’t not go!

4. Your research and that of your group covers the investigation and development of advanced novel statistical methodology driven by applications in the life, clinical, physical, chemical, engineering and ecological sciences. What are you working on currently?

I think the main thing I’m really working on at the moment is due to a post I’ve taken up at the Alan Turing Institute. Again, I’ve always been interested in engineering, so at the moment, combining statistics with hard core engineering and applied mathematics is for me just absolutely fantastic.

The Institute is just getting up and running. It’s really nice. They’ve got a really beautiful space, and it’s located on the first floor of the British Library.

5. As well as your current role at Imperial, you are also Director of Data Centric Engineering Programme at The Alan Turing Institute. Please could you tell us more about this role and the work that goes on at the Institute?

The role is such that I’m leading this program which is funded by the Lloyd Register Foundation to 10 million, so it’s quite substantial, and the idea that is described in a technology foresight review which the Lloyd Register Foundation commissioned about the impact Big Data would have on engineering and how we could make engineering infrastructure, products and services safer. So I spent half of my time at the Turing Institute building this program up and the other half at Imperial. So it’s very exciting. I’m working with lots of companies, ranging from big companies like Shell to small startups. Working with government departments, chief scientific advisors, and I’m really defining this whole discipline that’s called data-centric engineering. It’s early days but it’s very exciting.

6. What do you think have been the most important recent developments in engineering and will these influence your work in future years?

One of the major things has been what’s called additive manufacturing, where you basically print in 3D an object, rather than taking a solid block of material and cutting it down. So from an engineer’s perspective this is very exciting, because it’s a whole way of fashioning artifacts, engineering components, and so forth. From a statistician’s perspective, it’s very exciting. If you work with stainless steel, you know what its properties are. The processes that are used to roll and to create stainless steel are very well known, and so you can control the properties to very high degrees of tolerance. With additive manufacturing, the amount of variability in the properties of material is completely unknown, and it’s highly stochastic, it’s very random, and consequently that means that an awful lot of these engineering problems—design of parts that are made with metal and the actual certification that these things are safe and what their long-term properties are, how are they going to perform on the roadunder loading—these are now very much statistical questions. So there’s going to be a huge amount of work that needs to be done in additive manufacturing from a statistician’s and an engineer’s perspective.

7. Your research has been published in journals and books: is there a particular article or book that you are most proud of?

I think that the article I’m most proud of is a paper that was published in the Journal of the Royal Statistical Society Series B, which was selected to be presented before the Royal Society, so it was what’s called a “read paper”—a paper with discussion. That paper got the most number of contributions to the discussion than any paper that was ever presented to the Royal Statistical Society in its 180-odd year history. People like David Cox and C.R. Rao contributed to the discussion, so that makes me very proud of that and the paper itself has actually had some quite serious impact.

8. What is the best book in statistics that you have ever read?

I would say that the books that really have an impact on me in statistics are ones about the history of statistics. Reading about Fisher and his life as a scientist, and you’re reading about Pearson and reading about all the greats in statistics, I thought for me is something that I find really inspiring.

9. What would you recommend to young people who want to start a career in statistics?

I think get the foundations absolutely right. Do a good load of maths courses and a good load of fundamental statistics courses. I think that anyone that wants a career in statistics needs to understand the underlying mathematics and needs to have that intuitive feel for data and what it might be telling you. I think the third thing really important thing would be to learn to program and program to a very high level of expertise. Then you can call yourself a Data Scientist as well as a Statsitician !

10. Who are the people who have been influential in your career?

I think everyone that’s hired me has been influential in my career! So some of the people that I hugely respect are ones that hired me.


Copyright: Image appears courtesy of Professor Girolami