Abstract Details
Activity Number:
|
433
|
Type:
|
Invited
|
Date/Time:
|
Wednesday, August 6, 2014 : 8:30 AM to 10:20 AM
|
Sponsor:
|
Section on Statistical Learning and Data Mining
|
Abstract #310651
|
View Presentation
|
Title:
|
Fast Wiener Filtering and the Bayesian Lensing Challenge
|
Author(s):
|
Ethan Anderes*+
|
Companies:
|
University of California, Davis
|
Keywords:
|
Wiener filtering ;
Cosmic Microwave Background ;
Bayesian lensing
|
Abstract:
|
Estimating the gravitational lensing of the Cosmic Microwave Background (CMB) has become an exciting new avenue for probing the nature of dark matter and constraining cosmological models of universe. Nearly all of the recent statistical results on gravitational lensing utilized the, so called, quadratic estimator developed by Hu and Okamoto (2001, 2002). However, the results of Hirata and Seljak (2003) demonstrate that the quadratic estimate based on the polarization fields can be significantly sub-optimal to the maximum likelihood estimate. Unfortunately likelihood inference becomes computationally prohibitive when masking and nonstationary noise corrupt the observations. In this talk we present some new techniques that circumvent some of these computational challenges. We present two new FFT based algorithms which make an approximate Gibbs sampling from the posterior feasible. One of the algorithms uses a message passenger approach which is specifically designed to handle large masking and nonstationary noise for fast conditional simulation of the de-noised lensed CMB field. The second algorithm utilizes a FFT characterization of the gradient of the likelihood with respect to th
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2014 program
|
2014 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Professional Development program, please contact the Education Department.
The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Copyright © American Statistical Association.