Dr. Wood (c.v.) is a statistical machine learning researcher.
New
- Low rank continuous-space graphical models
- Inference in Hidden Markov Models with Explicit State Duration Distributions
- M. Dewar, C. Wiggins, and F. Wood, IEEE Signal Processing Letters, 2012 .bib
Cool
- W. Neiswanger made this multiple object target tracking video for a paper we’re about to submit.
- J. Huggins made a simple "visualization" of the writing of a recent NIPS paper entitled Hierarchically Supervised Latent Dirichlet Allocation.
Highlights
- The Sequence Memoizer
- F. Wood, J. Gasthaus, C. Archambeau, L. James, Y.W. Teh, CACM, 2011. .bib, Talk, Video Lecture, Code
- Probabilistic Deterministic Infinite Automata
- D. Pfau, N. Bartlett, F. Wood, NIPS 2011. .bib, spotlight video
- Forgetting counts : Constant Memory inference for a dependent hierarchical Pitman-Yor Process
- A hierarchical nonparametric Bayesian approach to statistical language model domain adaptation
Philosophy
Dr. Wood builds 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). He is particularly interested in pursuing the connections between specific features of Bayesian machine learning algorithms and the biological mechanisms they very much resemble.