Faculty Spotlight: David Gomez-Ullate

David Gomez-Ullate is more than a popular professor of Applied Mathematics at IE’s School of Science and Technology. He also packs strong research credentials. His early papers on the theory of exceptional orthogonal polynomials have received international prizes and belong to the 0.2% of most cited papers in mathematics. 

He has been a plenary speaker at the main conference in this field: Orthogonal Polynomials, Special Functions and Applications (OPSFA), presenting work at seminars in Cambridge, Harvard, Rome, Stockholm, Mexico, Copenhagen, India, and Hong Kong, among others. The BBVA Foundation strongly supports his career thanks to the Leonardo Scholarship award (2015), Research Groups in Big Data (2020) and the current project he leads for optimization of maritime routes – for which his teams have recently won prizes in competitions such as the Ocean Hackathon and the 2021 OpenCV Competition.

Nonetheless, Gomez-Ullate is strongly committed to the formation of younger scientists, with 15+ years of teaching. He has supervised more than 25 students towards PhD, MSc and BSc projects. Passionate about science dissemination, he has written a book on Big Data, delivered public lectures at numerous events, written articles for mainstream Spanish newspapers (El País, ABC), and participated in interviews and round tables for media.

Meet one of IE School of Science and Technology’s thought leaders:

What brought you to IE? Where were you before? And what keeps you here?

Most of my academic career has been linked to Complutense University, my alma mater. I have occupied full-time research positions in Canada (McGill and CRM), Italy (Università di Bologna) and Barcelona (UPC) and visiting positions in France (Université de Lorraine) and UK (University of Kent). Just before joining IE, I was leading a research datalab at the University of Cádiz. We assembled a great team of people and with scarce resources we carried out excellent research attracting a good number of projects.

Last year I found out that IE was launching a new degree in Applied Mathematics, and I heard about the growth plans for the School of Science and Technology. I felt this was an excellent opportunity to join this new exciting project (but I do miss watching sunsets over the ocean!)

Where do you work outside of IE and what is most exciting to you about this role in your professional career?

Besides my work at IE, I am leading a research project on optimization of maritime routes with weather and ocean data and forecasts. Together with some of my former PhD students we have founded a startup to develop this technology, which might have a considerable impact in decarbonizing the maritime shipping industry. Previously I had also founded a consulting technological company that specializes in data science and AI. I am always excited when I work on knowledge transfer, finding ways to apply my mathematical knowledge to solve real world problems.

Have you ever had an a-ha moment while teaching that furthered your research? What was it?

Asking out-of-box questions drives possible breakthroughs in mathematical research. Most of the time, these questions are just nonsense, and nothing comes out of them, a kind of mental divertimento. But sometimes, one of these “silly” questions leads to new insights and ideas that turn out to open entire new fields of research. Back in 2008 I began to question some of the well-established principles in the theory of classical orthogonal polynomials, which are central in the way we approximate functions in many fields of science and engineering. This led to the discovery of exceptional orthogonal polynomials, whose theory has been developed in the past 15 years by my group and many other researchers from all over the world. The a-ha moment when one visualizes how to solve a problem, or the idea that will help to complete a mathematical proof usually comes after many hours of hard concentration and deep dive into the problem. However, enlightenment does not hit you while you are working on it, but often a short while after when you are doing something else, as if your brain kept working on the background without you noticing. These moments are very very rare… and incredibly pleasant!

What book do you wish your students would read before taking your class and why?

I am currently teaching Deep Learning, which evolves so rapidly that by the time a book goes into print, a good part of its contents are already outdated. The book Deep Learning by Goodfellow, Bengio and Courville is a standard reference and does a great job in laying down the main concepts and ideas. It is a reference book though, I think few people will read it from start to end. For instance, the book was published just before the transformer architecture was discovered, and now transformers are state-of-the-art in almost every complex ML task. Most technical contents are transmitted via research papers and blogs, there’s no time to publish books on this matter. A related book that was a bit of an eye opener to me was Weapons of math destruction by Cathy O’Neill. It was the first time I reflected on “the dark side” of AI and data science. Some of these topics now are part of the public debate, but when the book was first published in 2016, she was one of the first people to bring attention to potentially dangerous social implications of these technologies.

Please name one of your articles or studies you feel addresses the most important issues for 2023?

During the pandemic years, society turned its attention to scientists seeking for understanding and solutions to a global problem. It was an incredibly active period. I had the chance to participate in an experts panel of mathematicians selected by CEMat (the ensemble of all spanish mathematical societies) to advise and work hand in hand with epidemiologists from CCAES, who elaborated the main recommendations to the government. The main question then was which of all the possible non-pharmaceutical interventions (NPIs), i.e. restrictive measures, were the most efficient ones in fighting against the spread of coronavirus. We gathered a large number of data and defined stringency indices for each province in Spain, and created statistical models to assess the efficiency of each measure. The results are displayed in this website and the paper has recently been published: Effectiveness of non-pharmaceutical interventions in nine fields of activity to decrease SARS-CoV-2 transmission (Spain, September 2020–May 2021). Front. Public Health, 12 April 2023.

Whose research of your IE colleagues do you find interesting? Why?

There are many interesting initiatives being launched at IE around implications of AI in different fields, such as law, architecture and design, geopolitics and public affairs, and of course in other fields of business and economy. I have a closer knowledge my colleagues’ research at the Sci-Tech School, where we have recently created the IE Research Datalab. We are hoping very soon to be able to contribute with our deep tech knowledge to some of the fundamental research questions that other research groups at IE are doing around AI.

Tell us one personal thing about yourself that none of your students know. A hobby, sport or talent? Strange fact? Unusual interest?

I played rugby for many years before a number of knee injuries forced me to retire from the fields. It is a great sport with very strong team values and I think this experience has taught me so many positive things. Now I play tennis and golf and I love to cycle and hike in the mountains. I used to be a fairly decent cook, when I had more time to devote to it. I like to search for mushrooms and I can identify quite a few (this was more fun before AI) but I only eat a small fraction of them. My biggest passion is to spend time with my wife and daughters.