Duke Institute for Brain Sciences

I am interested in developing new statistical methods using hierarchical and nonparametric Bayesian models, as well as other techniques from Bayesian statistics and machine learning, for extracting useful information from data when little or no supervision is available. I apply these methods to problems in cognitive science, where I strive to model human behavior, including human categorization, perception, and social interactions in online environments.

NSF Postdoctoral Fellow, Massachusetts Institute of Technology, 2011-2012

EPSRC Postdoctoral Fellow, University of Cambridge, Engineering Department, 2008-2011

Ph.D., University College London (UK), Gatsby Computational Neuroscience Unit, Machine Learning, 2008

M.S., Columbia University, Computer Science, 2003

B.S., State University of New York at Stony Brook, Applied Mathematics and Statistics, Computer Science, 2000

J. T. Abbott, K. A. Heller, Z. Ghahramani, and T. L. Griffiths. Testing a Bayesian measure of representativeness using a large image database. In Neural Information Processing Systems (NIPS), 2011.

R. Silva, K.A. Heller, Z. Ghahramani, and E. Airoldi. Ranking Relations using Analogies in Biological and Information Networks. In Annals of Applied Statistics (AOAS), 2010.

K.A. Heller, A. Sanborn, and N. Chater. Hierarchical Learning of Dimensional Biases in Human Categorization. In Neural Information Processing Systems (NIPS), 2009.