Ted Meeds
I am a Machine Learning engineer with a wide breadth of experience. Modeled data from financial sources (high-frequency tick, analysts' recommendations), bioinformatics, documents, computer vision, industrial manufacturing.
 
My PhD research focused on the application nonparametric Bayesian priors to novel machine learning models.
 
I am currently developing web-based machine learning applications and I am interested in moving out on the finance world and into Big Data (personalization, recommendation, bioinformatics, etc).  
Publications
 
Edward Meeds,
Nonparametric Bayesian Methods for Extracting Structure from Data,  
PhD Thesis, Department of Computer Science, University of Toronto, 2008. [pdf]
 
Edward Meeds, David Ross, Richard Zemel, and Sam Roweis,
Learning Stick-figure Models using Nonparametric Bayesian Priors over Trees,
Computer Vision and Pattern Recognition, 2008 (to appear). [pdf]
 
Edward Meeds and Sam Roweis,
Nonparametric Bayesian Biclustering,
UTML-TR-2007-001, Technical Report, University of Toronto, 2007. [pdf]
 
Edward Meeds, Zoubin Ghahramani, Radford Neal, and Sam Roweis,  
Modeling Dyadic Data with Binary Latent Factors,  
Neural Information Processing Systems,  2006. [pdf]
 
Edward Meeds and Simon Osindero,
An Alternative Infinite Mixture of Gaussian Process Experts,
Neural Information Processing Systems, 2005.  [pdf]
 
Edward Meeds,
Novelty Detection Model Selection Using Volume Selection,
UMTL-TR-2005-004, Technical Report, University of Toronto, 2005. [pdf]
CV: [pdf]
email: tmeeds@gmail.com