Professor Afonso Bandeira is an Assistant Professor of Mathematics at the Courant Institute of Mathematical Sciences. This fall, he will be teaching a course at the Center for Data Science, titled, “Optimization and Computational Linear Algebra for Data Science.”
What did you study in school?
I did my Bachelors of Science and Masters of Science in Mathematics, at the University of Coimbra in Portugal, and I became interested in applied mathematics while studying the connections between harmonic analysis and signal processing. I later completed my Phd in Applied and Computational Mathematics at Princeton University. Since then, I have studied the mathematical processes behind data extraction.
Could you tell me about some of the research projects that you’re working on?
The cryo-electron microscopy problem is a good example of the type of problem I tend to research. Say you have a projection of an unknown molecule structure, and that projection is given from an unknown direction. If you knew one of the variables—either the direction of the projection, or the structure—then you could probably deduce the other unknown variable. But with the cryo-electron microscopy problem, neither variable is known, so determining both at the same time becomes an interesting mathematical problem.
Can you talk about the course that you’re teaching in the fall?
The course I will be teaching is titled, “Optimization and Computational Linear Algebra for Data Science.” The course aims to provide students with the optimization and linear algebra skills that are commonly needed in the broader field of data science.
How do optimization and linear algebra fit into a broader data science curriculum?
While a lot of data science techniques and methods—machine learning, neural networks, artificial intelligence—are continuing to evolve, the principles of optimization and linear algebra will essentially remain the same, and will continue to be foundational to data science in the coming decades.
What drew you to the Center for Data Science at NYU?
There’s a level of excitement in being at a place where such a diverse set of research backgrounds are united by an interest in data-driven and data-inspired research. That sort of environment creates potential for collaborations across fields that would, otherwise, potentially not interact.
On a chalk board at CDS, we have a question: what does it mean to be a data scientist? What does that mean to you?
For me, being a Data Scientist is not only learning from data itself, but also understanding the process—both the potential and the limits—of learning from data, and aiming to develop methods and tools that take advantage of data’s full potential.