Professor Computer Science
Tony directs the Columbia Machine Learning Laboratory. Our research area is machine learning, a field which develops novel algorithms that use data to model complex real-world phenomena and to make accurate predictions about them. Our work spans both the applied and the fundamental aspects of the field. We have made contributions by migrating generalized matching, graph-theoretic methods and combinatorial methods into the field of machine learning. We have also introduced unifying frameworks that combine two separate methodologies in the field: discriminative approaches and generative approaches. The machine learning applications we have worked on have had real-world impact achieving state of the art results in vision, face recognition and spatio-temporal tracking, leading to a successful startup which applies machine learning to spatio-temporal data. Our papers are available online here.