0.25
0.5
0.75
1.25
1.5
1.75
2
Fully Bayesian Source Separation with Application to the CMB
Published on 2008-02-154175 Views
Blind source separation refers to the inferring of the values of variables (known as sources) from observations that are linear combinations of them. The observations and sources are usually vectors.
Presentation
E-Team on Unsupervised Segmentation Fully Bayesian Source Separation with Application to the Cosmic Microwave Background00:00
The Problem - Factor Analysis06:58:53
Cosmic Microwave Background (CMB)19:52:35
CMB Spectrum - Black Body35:51:11
Inferring the CMB - Source Separation39:44:39
Separating the Cosmic Microwave Background51:30:14
Model77:13:05
Model for Sources84:13:43
Model for Mixing Matrix A - 1101:10:54
Model for Mixing Matrix A - 2127:11:33
Priors145:58:45
Sampling from the Posterior Distribution161:17:36
Example 1: Simulated Data - 1178:43:01
Example 1: Simulated Data - 2199:46:04
Example 1: Simulated Data - 3204:30:06
Example 1: Simulated Data - 4208:24:39
Example 1: Simulated Data - 5214:34:51
Example 1: Simulated Data - 6224:35:39
Example 1: Simulated Data - 7232:04:50
Example 1: Simulated Data - 8239:52:33
Example 1: Simulated Data - 9244:45:04
Example 2: Simulated Data with 9 Channels - 1244:59:43
Example 2: Simulated Data with 9 Channels - 2253:52:19
Example 2: Simulated Data with 9 Channels - 3256:54:57
Example 2: Simulated Data with 9 Channels - 4257:11:07
Example 2: Simulated Data with 9 Channels - 5258:26:18
Example 2: Simulated Data with 9 Channels - 6263:35:35
Real WMAP Data263:47:49
Patch 2 - Data280:34:57
Patch 2 - Posterior Mean of Sources286:46:00
Patch 2 - Posterior Standard Deviation of Sources290:52:12
Patch 2 - Model Fit: Observed Temperature vs. Posterior Mean Temperature296:42:18
Patch 3 - Data307:52:39
Patch 3 - Posterior Mean of Sources309:56:07
Patch 3 - Posterior Standard Deviation of Sources312:07:49
Patch 3 - Model Fit: Observed Temperature vs. Posterior Mean Temperature314:38:47
Posterior Mean of Spectral Indices316:34:40
Future Work327:38:38