How would you describe your work in 90 seconds?

The main goal of my work is to provide the basis for intelligent techniques that are reliable, precise, and trustworthy, and that we can understand because they’re based on knowledge and reasoning, on the knowledge that we have, and on trying to simulate our reasoning. I want to understand how we can make that possible and what the main challenges are in this area. This includes questions like how hard is it to find accurate information of a certain type for a specific domain, how can we find the correct answers if we organize it in a specific way, and what if our data is incomplete or of low quality? How can we still find precise answers to our information needs? How can we draw accurate inferences that we can rely on and that we know are correct? So, how can we get to reliable, high-quality, intelligent techniques? My work is also about understanding what the computational challenges are when it comes to those questions and overcoming them. How can we have algorithms that run efficiently, and that quickly get us the answers to our questions? That’s the main underlying theme of my research.

How did you get in touch with informatics?

It’s funny because I was not considering doing computer science as such. I was always very keen on math, and when I was in high school, I was in competitions like the Mathematics Olympiad. I discovered that I like the concept of mathematical proofs, and being able to prove that something is correct. I found it fascinating, and I wanted to do something with a lot of math, and especially with mathematical proofs. I didn’t know that computer science could offer me that. At the same time, when I was a teenager in Mexico, I was also very involved in social projects. I was working with street children; I was teaching math to adults and blind people, and I felt that studying math was too abstract. I wanted to do something useful for humanity, and I wasn’t sure that pure math would be the right answer. So, my compromise was to look for a discipline with a lot of math that could also be applied. By looking at the programs of different engineering faculties, I saw that computer science had a lot of math. I thought, computers are useful, so maybe this could be a reasonable compromise between these two interests that were both pulling in opposite directions. What I didn’t know back then was that it was a perfect compromise for me because I discovered that the math that we do in computer science, discrete math, and especially the role of logic in computer science, was just right for me. I have always been fascinated by logic and the ability to use it, to understand what is correct, and to figure it out with precision. I found that fascinating, and I also discovered that programming was super cool. I was lucky to land in the right place, and one lesson I learned from this story that brought me to logic and to a job that I really like is that we have very different talents. Some things are more clearly and easily applicable than others, but in the end, the one thing that we can do is to do what we are good at in the best possible way and to work together.

Where do you see the connection between your research and everyday life?

Logic in computer science can have a huge impact; it allows us to have systems that are reliable, that are trustworthy, that we can communicate with. Artificial intelligence is not going away, it’s present in every aspect of our lives, but there are many systems that are obscure. We don’t know what is happening inside of them; we don’t know what data they were trained on, and we don’t know if they contain errors. We have all these problems with hallucinations, for example. For many of our problems, our everyday information needs, and for decision-making processes that are necessary in many areas, we shouldn’t be guessing. We actually have tools that allow us to have understandable, precise, reliable, intelligent techniques. That’s what I want to make possible with my research; it just so happens that one can do it with fancy logic. I really believe that symbolic Artificial Intelligence can, in many cases, provide us with better solutions, and we shouldn’t settle for less. We can have solutions that we can trust, that will not be biased, and that we can understand. Even if there is an error, we can find out what it is and why it happened, and not just hope that a system will give us the correct answer.

What makes you happy in your work?

To be able to concentrate on a hard problem for several days and find a solution will never lose its charm, for me. I also really enjoy teaching. Seeing students discover what they like, discover what they’re good at, and helping them to understand and learn. Teaching logic, and as a supervisor, helping a student find a problem that they’re interested in and trying to solve a real-life problem with mathematical techniques also makes me happy. Now that I’m working with students, it’s also very nice to be able to share this passion with other people.

Why do you think there are still so few women in computer science?

I think a large part of it is social roles that we have been following for so many years. It was only a little over 100 years ago that women couldn’t even go to university, and now we have female professors. We are moving in the right direction, but there is still a lot to do. The power structures have been biased towards men for a long time, and so have the gender roles that we still enforce. We make girls believe from very early on that computer science and technology are for boys, and that language and people are for girls. We lose a lot of talent that way, and I’m happy that I have so many excellent female students and postdocs. My research group is not balanced in terms of gender - we are more women than men - but that’s not the case in most parts of the faculty. I think it’s important to have role models and to help girls discover that math is cool and that technologies are also an option for them. We have to keep encouraging young girls, but then there are also structural factors. There are still many things in academia that contribute to this problem: You are expected to travel a lot, there are part-time contracts, and you’re expected to move many times. This system was built under the assumption that professors would be men whose wives don’t work and take care of the children. This is something that I still struggle with. To some extent, it’s expected that you can show up to a workshop on a Sunday or during holidays, or stay for meetings after 6 p.m. This mindset at the very basis of academic environments reflects that of male professors with stay-at-home wives, which is very challenging when you try to follow the same path as a mother, as a woman with a family, and that can be discouraging. That also contributes to this leaky pipe. Even the girls who do start computer science at some point realize that if they want to have a family, it will be tough. That’s why we should show them that it’s possible, but we should also make it easier, because there are many aspects of academia that are still not women-friendly, not family-friendly, and universities as organizations are still biased against certain groups in that regard.

Magdalena Ortiz is a Professor of Knowledge-Based Systems at TU Wien Informatics. Magdalena Ortiz will hold a joint Inaugural Lecture on October 15, 2024; 17:00 CEST at EI 9, Gußhausstraße together with Emanuel Sallinger and Dominique Schröder.

Discover the whole #5QW series.

Curious about our other news? Subscribe to our news feed, calendar, or newsletter, or follow us on social media.