01.04.2021
| Inside Data Science |
Norman Sieroka on Philosophy and Data Science
Norman Sieroka is Professor of Philosophy and deals with philosophical questions related to the handling of data and cognitive processes.
What topics are you currently working on in your research?
I deal a lot with the topic of time in my research. Due to my background in physics, I am interested in physical time on the one hand: How do we measure time? How do we determine time in physics? But on the other hand, I am also concerned with questions about the experience of time, for example related to phenomena such as boredom or even changes that we experience as a result of Corona. The second major area of my research is the philosophy of individual scientific disciplines: How do sciences work? How do we gain knowledge? For these questions, the handling and understanding of data is particularly important. This is where I am interested in the context of the exact disciplines – including mathematics and computer science in addition to physics. I also have projects with colleagues from pharmacy, history, and architecture.
How important is data to your research?
Philosophy is a reflective discipline. Thus, in the literal sense, we are dependent on “data”, on something given or predetermined. Philosophers themselves typically do not collect empirical data in their research; the data of philosophy are actually books and texts.
And then there are concrete research projects in which I deal with philosophical questions about data and data handling. For example, we just won a project on “AI augmented architectural design” with colleagues from Zurich. This project explores the use of AI methods in design processes, for example, to create walls or facades with certain properties. And indeed, this also raises a number of exciting philosophical questions.
What role does data science play in your research? Do you see yourself more as a user, a method developer, a basic researcher, or perhaps something completely different?
I would not describe myself as a basic researcher in the strict sense, but in the broader sense, since I deal with fundamental questions from a reflective point of view. So, for me different approaches in different disciplines and research contexts are very interesting and insightful. For example, while one discipline is characterized by the advancement of existing theories, another focuses primarily on concrete and detailed solutions to problems. But both work data-intensively in this regard and do so with partly similar, partly different methods. In addition, data or methods of data analysis also play an important role in the context of my research on the topic of time. Here one quickly comes to questions about the connection between temporal correlations and causal dependencies – keyword “causal interference”. This is also a highly topical issue in the context of the COVID-19 pandemic – one might think, for instance, of the discussion about vaccine side effects: How can we infer causal dependencies from isolated cases of thrombosis, for example?
Which data science methods and technologies are in the focus of your research or could also become interesting in the future?
Regarding the question of how cognitive processes change, I am interested, for example, in what is outsourced to the computer or the robot, so to speak, and what is not. I’m also interested in when one would describe results obtained with the help of AI as genuinely new or even creative. Or also the question of the relationship between automation and personalization in design processes – a question that can be applied not only to architecture, but also in particular to drug design and personalized medicine.
A fundamental question that arises overall is the relationship between predictions, which are becoming better and more successful on the basis of data, and scientific understanding or theory building. There are hopes as well as concerns: predictions are becoming more accurate, but aren’t we conversely neglecting scientific insights or explanations? Or does this concern become redundant with successful implementation? If, for example, neuroscientists would build an artificial retina tomorrow that works perfectly, then one could say that the question of a theory of vision would become obsolete. After all,“obviously” everything essential has already been implemented. To put it in a philosophical slogan: verum factum – only what we ourselves have created is true anyway. – Or is it not?
What are your main challenges in dealing with data?
For me, the great challenge is to better understand the general handling of data and, in particular, the dynamics and cognitive processes behind it or made possible by it. The basic idea of such a philosophical reflection is the following: The better I understand something, the better prepared I am for what is to come. I may not have a one-to-one solution to all problems, but I generally have a better understanding of how data is handled in different contexts and can thus also better adapt to new situations. And of course, these new situations are also linked to many practical questions regarding the handling of data – for example, questions about data security, data protection, or international agreements such as the Nagoya Protocol. In my view, however, what we need here above all is a theoretical understanding so that we can then address these practical issues in a meaningful way.
And finally, what is your personal motivation for joining the Data Science Center?
Interdisciplinarity and reflection. On the one hand, I actually grew up interdisciplinary myself: since the first semester, I have studied philosophy, physics and math. Accordingly, I very much appreciate interdisciplinary centers and cross-disciplinary platforms at the university and I am happy about the collaboration and exchange in the Data Science Center. In fact, the exchange with other researchers who work with data or use AI is a content-related prerequisite for being able to investigate philosophically relevant aspects at all.
Please note: The interview was originally given in German and translated into English by Lena Steinmann.
You can learn more about Norman Sieroka’s activities in his talk
“Data Science and Philosophy“ in the Data Science Forum on 08.04.2021.
« back
01.04.2021 | Inside Data Science
Norman Sieroka on Philosophy and Data Science
Norman Sieroka is Professor of Philosophy and deals with philosophical questions related to the handling of data and cognitive processes.
What topics are you currently working on in your research?
I deal a lot with the topic of time in my research. Due to my background in physics, I am interested in physical time on the one hand: How do we measure time? How do we determine time in physics? But on the other hand, I am also concerned with questions about the experience of time, for example related to phenomena such as boredom or even changes that we experience as a result of Corona. The second major area of my research is the philosophy of individual scientific disciplines: How do sciences work? How do we gain knowledge? For these questions, the handling and understanding of data is particularly important. This is where I am interested in the context of the exact disciplines – including mathematics and computer science in addition to physics. I also have projects with colleagues from pharmacy, history, and architecture.
How important is data to your research?
Philosophy is a reflective discipline. Thus, in the literal sense, we are dependent on “data”, on something given or predetermined. Philosophers themselves typically do not collect empirical data in their research; the data of philosophy are actually books and texts.
And then there are concrete research projects in which I deal with philosophical questions about data and data handling. For example, we just won a project on “AI augmented architectural design” with colleagues from Zurich. This project explores the use of AI methods in design processes, for example, to create walls or facades with certain properties. And indeed, this also raises a number of exciting philosophical questions.
What role does data science play in your research? Do you see yourself more as a user, a method developer, a basic researcher, or perhaps something completely different?
I would not describe myself as a basic researcher in the strict sense, but in the broader sense, since I deal with fundamental questions from a reflective point of view. So, for me different approaches in different disciplines and research contexts are very interesting and insightful. For example, while one discipline is characterized by the advancement of existing theories, another focuses primarily on concrete and detailed solutions to problems. But both work data-intensively in this regard and do so with partly similar, partly different methods. In addition, data or methods of data analysis also play an important role in the context of my research on the topic of time. Here one quickly comes to questions about the connection between temporal correlations and causal dependencies – keyword “causal interference”. This is also a highly topical issue in the context of the COVID-19 pandemic – one might think, for instance, of the discussion about vaccine side effects: How can we infer causal dependencies from isolated cases of thrombosis, for example?
Which data science methods and technologies are in the focus of your research or could also become interesting in the future?
Regarding the question of how cognitive processes change, I am interested, for example, in what is outsourced to the computer or the robot, so to speak, and what is not. I’m also interested in when one would describe results obtained with the help of AI as genuinely new or even creative. Or also the question of the relationship between automation and personalization in design processes – a question that can be applied not only to architecture, but also in particular to drug design and personalized medicine.
A fundamental question that arises overall is the relationship between predictions, which are becoming better and more successful on the basis of data, and scientific understanding or theory building. There are hopes as well as concerns: predictions are becoming more accurate, but aren’t we conversely neglecting scientific insights or explanations? Or does this concern become redundant with successful implementation? If, for example, neuroscientists would build an artificial retina tomorrow that works perfectly, then one could say that the question of a theory of vision would become obsolete. After all,“obviously” everything essential has already been implemented. To put it in a philosophical slogan: verum factum – only what we ourselves have created is true anyway. – Or is it not?
What are your main challenges in dealing with data?
For me, the great challenge is to better understand the general handling of data and, in particular, the dynamics and cognitive processes behind it or made possible by it. The basic idea of such a philosophical reflection is the following: The better I understand something, the better prepared I am for what is to come. I may not have a one-to-one solution to all problems, but I generally have a better understanding of how data is handled in different contexts and can thus also better adapt to new situations. And of course, these new situations are also linked to many practical questions regarding the handling of data – for example, questions about data security, data protection, or international agreements such as the Nagoya Protocol. In my view, however, what we need here above all is a theoretical understanding so that we can then address these practical issues in a meaningful way.
And finally, what is your personal motivation for joining the Data Science Center?
Interdisciplinarity and reflection. On the one hand, I actually grew up interdisciplinary myself: since the first semester, I have studied philosophy, physics and math. Accordingly, I very much appreciate interdisciplinary centers and cross-disciplinary platforms at the university and I am happy about the collaboration and exchange in the Data Science Center. In fact, the exchange with other researchers who work with data or use AI is a content-related prerequisite for being able to investigate philosophically relevant aspects at all.
Please note: The interview was originally given in German and translated into English by Lena Steinmann.
You can learn more about Norman Sieroka’s activities in his talk
“Data Science and Philosophy“ in the Data Science Forum on 08.04.2021.
Interviewee:
Prof. Dr. Dr. Norman Sieroka
Professor of Philosophy
FB 09 – Cultural Studies
sieroka@uni-bremen.de
« back