Monday, April 11, 2016 - 4:00pm
Abstract: Machine learning has transformed the technology industry over the last decade, forming the basis for web search, speech recognition, product recommendations, and self-driving cars. With the increased adoption of electronic health records and a surge in funding for health IT startups, health care is undergoing a similar transformation. I will talk about a few of my group's recent advances in machine learning that have the potential to have a major impact in health care. The talk will focus in particular on a new approach to learning from temporal data, coupling deep learning with probabilistic inference. Applied to learning disease progression models from clinical data, our algorithms learn a rich representation that is capable of answering counterfactual questions such as which treatment is most appropriate to which patient, providing a new theoretical framework for precision medicine. Bio: David Sontag is an Assistant Professor at New York University's Courant Institute of Mathematical Sciences and Center for Data Science. His research is on machine learning and approximate inference in graphical models, with an emphasis on applications to health care. Prior to joining NYU, he was a postdoctoral researcher at Microsoft Research New England. David's research has received recognition including the Sprowls Award for the best doctoral thesis in Computer Science at MIT, best paper awards at EMNLP, UAI and NIPS, and the NSF CAREER award.