Interview “Extending the Limits of Our Curiosity With the Help of Technological Tools.”
Interview with Prof. Zeynep Akata, Director of the Institute for Explainable Machine Learning, Helmholtz Munich.
Interview with Prof. Zeynep Akata, Director of the Institute for Explainable Machine Learning, Helmholtz Munich.
Prof. Zeynep Akata's research focuses on developing explainable machine learning models for vision and language tasks. From building an institute from scratch to fostering creativity and tackling challenges in the fast-paced world of machine learning, Prof. Zeynep Akata shares her insights, aspirations and personal journey in this compelling interview.
Since when are you a Director at Helmholtz Munich?
ZA: I started in the position as the Director of the Institute for Explainable Machine Learning on January 1st 2024. My institute is a part of both the Computational Health Center and Helmholtz AI. In addition to my role at Helmholtz Munich, I am also a Liesel Beckmann Distinguished Professor of Computer Science at the Technical University of Munich, leading the chair of Interpretable and Reliable Machine Learning.
What inspires you for this position?
ZA: The most exciting aspect is being a part of two world class institutions such as Helmholtz Munich and TUM. This opens up several collaboration opportunities from different perspectives, for example applications domains, via my role at Helmholtz, also opportunities for influencing a larger range of audience profiles, e.g. teaching, student supervision, via my role at TUM. Now, it doesn’t feel like there is any considerable limit for my personal and professional growth. This process so far has not been an easy one but I am looking forward to tackling this challenge and observing the growth of the seeds that I’ll be planting over the years.
“I am looking forward to tackling this challenge and observing the growth of the seeds that I’ll be planting over the years.”
Prof. Zeynep Akata
Helmholtz MunichWhat does it mean to you to be selected by the 'Capital' Magazine as one of the 'Top 40 Under 40'?
ZA: This award is one of its kind for me as it is not only awarded to researchers or scientists but also people with any other role that shapes our society could be in this list. As a scientist, it is my wish to be useful for the society as a whole and not only influence my scientific community but the society as a whole. In that sense, it is a great honor to be recognized by the Capital Magazine as one of the "Top 40 Under 40".
What do you want to archieve with your research?
ZA: In fact, my entire research program is focusing on foundation models, for example models that are generalizable and adaptable to new tasks. At Helmholtz Munich, I am leading one of the few institutes that focuses on fundamental machine learning, without focusing on a particular application. This goes slightly against the usual practices of Helmholtz where the focus is on a particular application domain, like medicine, materials science and so on. Ever since the physics and chemistry Nobel prizes have been awarded to AI a few weeks ago, I believe that there is a shift in the society’s perception towards AI where people now do not question why foundation models and AI focused research should be funded. It should be funded because it is nowadays fundamental to any progress in the scientific domain. In that sense, I am proud to be one of the pioneers of this movement in Europe and I am ever more excited to work towards achieving foundation models that are generalizable and personalizable together with my students and colleagues.
Siarhei - stock.adobe.comWhat fascinates you about AI and foundation models?
ZA: AI agents are becoming more and more ubiquitous, more and more capable day by day. They can be thought of collaborators, agents that could work with us and help us. Anyone who has interacted with a conversational AI agent like ChatGPT must have experienced how capable these technologies are nowadays. I am grateful to be alive nowadays to experience this process during this boom. Now finally, we can set out to extend the limits of our curiosity with the help of the wonderful tools of technology, search and discover new horizons for impact, make a change and influence our society. This is truly fascinating, don’t you think?
What is special about Helmholtz Munich?
ZA: The domain knowledge of medicine at Helmholtz Munich fascinates me. If one would like to do research on cancer, diabetes, environmental health, or others, Helmholtz Munich is one of the best places to be, because of the access to data and access to talents. The more I discover Helmholtz Munich and all the domains that are being studied here, the more fascinated I get to be honest. I am in the process of getting to know the subjects being studied here and establishing collaborations with my colleagues which I am sure will open up new perspectives in my horizon of creativity.
Helmholtz Munich | Michael Haggenmueller“The more I discover Helmholtz Munich and all the domains that are being studied here, the more fascinated I get”
Prof. Zeynep Akata
What are your goals as the Director of the institute?
ZA: One of my goals is to build a safe space, a suitable environment where ideas can blossom, where young researchers would want to come, join us and fascinate us with their creativity. I would like to take part in creating a hub for machine learning researchers who are interested in applying their research findings to different application domains related to health and environment, make these technologies explainable such that they are more easily accessible to the user. My personal goal is to discover the different ways in which our research can be useful for the public good.
Gorodenkoff - stock.adobe.com“One of my goals is to build a safe space, a suitable environment where ideas can blossom, where young researchers would want to come, join us and fascinate us with their creativity.”
Prof. Zeynep Akata
What are the biggest challenges and why is it still worth it every day?
ZA: One of the biggest challenges is also one of the biggest advantages of working in my domain of machine learning (ML). ML models get more and more capable and more and more general in an extremely rapid pace. This opens tremendous opportunities in which one can make impact in the real world by just adapting existing methods to various data domains. These domains could be medicine, environment, earth observation, etc. It is very exciting and rewarding to discover so many new venues for research so rapidly. This is one of the aspects of ML that keeps me going. However, this fast-paced progress in technologies can cause constant stress in the workplace because there is always a risk of missing out so there is no time to waste. This affects not only me as a researcher but also my research staff so we are in a constant battle in keeping our research productive, our workspace psychologically safe and also everyone motivated.
Was there a formative experience in your career that left a mark on you?
ZA: : I have had many formative experiences in my career. I lived in Istanbul, Bonn, Grenoble, Saarbrucken, Berkeley, Amsterdam, Tübingen and now in Munich. This means I moved a lot, and I met many people, peers, supervisors, students who have all influenced me in their own ways. I had plenty of experiences concerning racism and sexism in my career, but I also had plenty of experiences feeling pleasure and joy doing my work. I got nervous, angry, felt helpless many times but I felt liberated, empowered as well. I feel lucky to have been given the chances that I had been given to succeed. While learning to survive and be successful, I also influenced junior people both positively and negatively, I am sure. My path until arriving in Munich wasn’t an easy one but now when I look back at my journey so far, I feel grateful that it has been an extremely interesting journey overall.
“My path until arriving in Munich wasn’t an easy one – but now when I look back at my journey so far, I feel grateful that it has been an extremely interesting journey overall.”
Prof. Zeynep Akata
How do you fuel your energy?
ZA: My family, especially my 2-year-old son, is a major source of happiness in my life. I believe that staying in the workforce, doing my research as best as I can sets an excellent precedent for him. This thought gives me strength and motivates me not only to do my duties at home but also at work. One hug from him means the world to me, it takes away all my stress and makes me forget my struggles.
Regrettably, I don’t have much time for hobbies, but I like going out for walks with my family or friends. I like the feeling of the wind or rain on my face, being in the nature in general. It doesn’t matter which activity I am engaged in as long as I am outside. We work indoors sitting all the time, so whenever I have the opportunity, I try to be outside. Munich and its surrounding have lovely spaces to offer for this, I feel lucky for living in this beautiful city.
Tell a secret about yourself!
ZA: My grandmother sent one of my drawings at the age of 3 to a national competition for children’s art in Turkey and I came first. Consequently, a picture of me during the prize ceremony with the other awardees appeared at a national newspaper. My mom kept a copy of the newspaper, and I can tell that the picture looks quite funny with me looking considerably more confused than the other two children.
Latest update: November 2024
Find Out More About Prof. Zeynep Akata & Related Research
Prof. Zeynep Akata is the Director of the Institute for Explainable Machine Learning at Helmholtz Munich. She is also a Liesel Beckmann Distinguished Professor at the Technical University of Munich. In November 2024, she was selected as one of the Top40 under 40 by the Capital magazine.
Contact: zeynep.akata@helmholtz-munich.de Profile: Prof. Zeynep Akata
