About me
I am a second-year graduate student, and when not doing problem sets, I spend my time working on analysis of type-1a supernovae found in the SDSS-2 Supernova Survey
Currently, I am putting most of my effort into working on algorithms for component-wise analysis of these supernova light curves, using variations of PCA and ICA. Right now I'm having a lot of success with Expectation-Maximization PCA to handle missing data. I'll post the code sometime soon...
update: I've begun exploring techniques in the field of Locally Linear Embedding and Hessian Eigenmaps. These are generalized nonlinear dimensionality reduction algorithms, which have possible application in the analysis of astronomical survey data.