About Me
I am a graduate student in computer science at the University of California, Irvine. I am currently in my third year of study, working with my advisers Alexander Ihler and Padhraic Smyth.
Still curious? Check out my CV or the work of my colleagues in the UCI Datalab.
Research Interests
I am interested in probabilistic graphical models as an elegant and effective way to deal with uncertainty in many real-world problems. Currently, their application is limited by the intractability of exact probabilistic inference and the lack of efficient methods for inference over continuous random variables.
I am working to extend the applicability of graphical models through improved approximate inference algorithms and efficient nonparametric methods for continuous models.
Publications and Code
A. Ihler, A. Frank, and P. Smyth. Particle-based variational inference for continuous systems. Neural Information Processing Systems, 2009. [pdf] [bib]
van Leeuwen, T. T., A. J. Frank, Y. Jin, P. Smyth, M. L. Goulden, G. R. van der Werf, and J. T. Randerson (2011), Optimal use of land surface temperature data to detect changes in tropical forest cover, J. Geophys. Res., 116, G02002, doi:10.1029/2010JG001488. [pdf] [bib]
Talks
Belief Propagation in a Continuous World - UCI AI/ML Seminar, 11/02/2009
Belief propagation is a popular algorithm for performing inference in probabilistic graphical models. It has been successfully applied in a diverse range of areas including computer vision, computational biology, and error correcting codes. Despite its many successes, however, the algorithm has weaknesses: it does not directly handle continuous random variables, and it may produce poor results on loopy graphical models. In this talk I will discuss several extensions to belief propagation that address these shortcomings. I will then show how these extensions can be combined to enable reasonable performance on loopy graphical models with continuous variables, with applications in localization and protein structure estimation. [ppt]