Manifold Regression via Brownian Motion

Manifold regression via Brownian Motion (MRBM) is a Bayesian regression algorithm that applies to the regression problem where the response takes manifold-value. Specifically, it estimates the function $f:[0,1]\rightarrow M$ where $M$ can be any general compact manifold. For the justification of the algorithm and its consistency theory, please refer to the paper below.

Matlab code for Manifold Regression via Brownian Motion.
Latest version of the code: MRBM.
Brief description of the algorithm can be found in the slides.


The material presented in this work is partially supported by NSF grants DMS-09-56072, DMS-14-18386 and Doctoral Dissertation Fellowship at UMN.



 

Back to Gilad Lerman's webpage


Last Modified Sunday April 23, 2017
The views and opinions expressed in this page are strictly those of the page author. The contents of this page have not been reviewed or approved by the University of Minnesota.