Kriste Krstovski

Kriste Krstovski is an Adjunct Assistant Professor at the Columbia Business School where he teaches PhD courses on machine learning and natural language processing. Additionally, he is an Associate Research Scientist at the Data Science Institute. Previously he was a postdoctoral research scientist at Columbia University and Yale University. Kriste received his PhD in Computer Science from University of Massachusetts Amherst. He spent the second half of his PhD as a Predoctoral Fellow at the Harvard-Smithsonian Center for Astrophysics where he was researching efficient latent variable models of text as part of the SAO/NASA Astrophysics Data System.

While at Harvard, Kriste was a runner[1]up in the Harvard President’s Innovation Challenge. Prior to starting his PhD studies, Kriste was a Staff Scientist in the Speech and Language Processing Department at Raytheon BBN Technologies. His research work at Raytheon BBN spanned various Defense Advanced Research Projects Agency (DARPA) ML and NLP projects which include DARPA’s TransTac, MADCAT and BOLT programs. While working full-time at Raytheon BBN, he attended two semesters at MIT as a special graduate student in the EECS department. Kriste finished his BS and MS in Electrical Engineering at the University of New Hampshire while being a member of the Consolidated Advanced Technologies for Law Enforcement Laboratory. Kriste was a key contributor in the development of the revolutionary, DoJ funded, Project54 system – a completely integrated voice[1]controlled police cruiser. Kriste’s research expertise spans across the fields of ML (probabilistic modeling, approximate Bayesian inference, and deep learning), NLP (Bayesian latent variable models of text, distributed representations, and machine translation), and information retrieval (recommender systems and multilingual document retrieval).