Highlights
- A Stochastic Memoizer for Sequence Data.
F. Wood, C. Archambeau, J. Gasthaus, L. James, and Y.W. Teh. ICML, 2009. PDF PDF Video Lecture
- A hierarchical nonparametric Bayesian approach to statistical language model domain adaptation.
F. Wood and Y.W. Teh. AISTATS 2009. PDF Best Paper Award
- A Nonparametric Bayesian Alternative to
Spike Sorting
F. Wood and M. J. Black, Journal of Neuroscience Methods, 173:1-12, 2008. PDF
Philosophy
My research effort is directed towards contributing models and algorithms to the field of statistical machine learning. My current research focus is on the development and application of nonparametric Bayesian methods to problems in natural language processing and compression.
I believe that the conceptual symbiosis between neuroscience and computer science (particularly in machine learning) is still in its infancy. As a computer scientist I believe that the computational superiority exhibited by even the most simple biological organisms is a strong argument for doing research at the intersection of these two fields.
My style is to build machine learning algorithms that improve on the state of the art by drawing from scientific findings at all levels of scientific investigation (physics to psychology). I am particularly interested in pursuing the connections between specific features of Bayesian machine learning algorithms and the biological mechanisms they very much resemble.