STATISTICS COLLOQUIUM
Thursday, May 13th, 2010 at 4.10pm, MSB 1147 (Colloquium Room)
Refreshments: 3.30pm, MSB 4110 (Statistics Lounge)
Speaker: Bin Yu (UC Berkeley)
Title: Sparse modeling: some unifying theory and understanding human visual pathway
Abstract: Information technology has enabled collection of massive amounts of data in science, engineering, social science, finance and beyond. Extracting useful information from massive and high-dimensional data is the focus of today's statistical research and practice. After broad success of statistical machine learning on prediction through regularization, interpretability is gaining attention and sparsity is being used as its proxy.
With the virtues of both regularization and sparsity, sparse modeling methods (e.g. Lasso) has attracted much attention for theoretical research and for data modeling.
In this talk, I would like to discuss both theory and practice of sparse modeling. First, I will present some recent theoretical results on bounding L2-estimation error (when p>>n) for a class of M-estimation methods with decomposable penalties. As special cases, our results cover Lasso, L1-penalized GLMs, grouped Lasso, and low-rank sparse matrix estimation. Second, I will present on-going research with the Gallant Lab at Berkeley on understanding visual pathway.
In particular, sparse models (selection by correlation, linear, non-linear) have been built to relate natural images to fMRI responses in human primary visual cortex area V1. Issues of model validation will be discussed.
