Welcome Nathan Kallus
Nathan Kallus has joined the faculty at the School of Operations Research and Information Engineering (ORIE) at Cornell Tech as an assistant professor. Kallus says he came to Cornell Tech in New York City “to participate in academic research that not only changes the way we think, but also how we do things, through direct engagement.”
Kallus was born in Haifa, Israel and lived there until he was 14. He then moved to the United States and took what amounted to a crash course in English as he enrolled in an American high school. Kallus did some computer programming in high school but by the time he went to college at Berkeley he had fallen in love with the beauty of mathematics and ended up earning a BA in mathematics in addition to a BS in computer science.
Kallus knew he wanted to go to graduate school and asked his undergraduate advisor, “What should I study if I know I want to solve interesting problems using math? And where should I go?” His advisor recommended he study operations research (OR) at MIT, so that is exactly what he did. After earning his Ph.D. from MIT, Kallus was a Visiting Scholar at USC’s Department of Data Science and Operations and a Postdoctoral Associate at MIT’s Operations Research and Statistics group. He came to Cornell Tech in 2016.
Kallus’s research revolves around data-driven decision making in operations, the interplay of optimization and statistics in decision making and inference, and the analytical capacities and challenges of observational, large-scale, and web-based data.
For example, Kallus has looked at how to make personalized diabetes management recommendations using electronic medical records. Current management guidelines put out by the American Association of Clinical Endocrinologists (AACE) are fairly agnostic to individual characteristics. Kallus is working on an algorithm that can take advantage of the massive amounts of data now available in electronic medical records to make personalized medicine really work for those with diabetes.
He has also done work with the Twitter stream to see if it could be used to predict events such as mass protests. Kallus analyzed tweets associated with the civil unrest that arose coincident with the coup d’etat in Egypt in 2013. He created a tool to sort through and analyze tweets and cross reference these with news media in order to predict the chances of future events.
Much of his work revolves around the problem of optimization under uncertainty. Kallus believes that most things in life cannot be known with certainty. But statistics can help us better understand things that are not absolute. “Optimization is the mathematics of making the best decision possible given a particular setting,” says Kallus. “An optimal decision in an unrealistic setting could be very bad in practice. So it is critical to recognize uncertainty when optimizing—including uncertainty in one’s model.” At Cornell Tech, Kallus plans to continue to do basic research that leads to novel applications that can have positive impact on the world.
Kallus lives in New York City with his husband who works for the British Government.