Figure 3 - available via license: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International
Content may be subject to copyright.
Download
View publication
Empirical stabilization. The peak at iteration 101 corresponds to the switch from the warm-up stage to the refinement stage.

Empirical stabilization. The peak at iteration 101 corresponds to the switch from the warm-up stage to the refinement stage.

Source publication
Figure 1: HEALPix discretization of the sphere (source: [13]). Moving...
Figure 2: Example of a Gaussian filterˆHνfilterˆ filterˆHν , with a...
Figure 3: Empirical stabilization. The peak at iteration 101...
Figure 5: Mean performance metrics over more than 30 realizations as a...
Joint deconvolution and unsupervised source separation for data on the sphere
Preprint
Full-text available
  • Dec 2020
Tackling unsupervised source separation jointly with an additional inverse problem such as deconvolution is central for the analysis of multi-wavelength data. This becomes highly challenging when applied to large data sampled on the sphere such as those provided by wide-field observations in astrophysics, whose analysis requires the design of dedic...
Cite
Download full-text

Context in source publication

Context 1
... toward a local minimum (e.g. BCD [24], PALM [25]). SDecGMCA is built upon a projected ALS scheme. To the best of our knowledge, is has not been theoretically proved that projected ALS algorithms converge. However, we empirically show that the regularization parameters tend to stabilize along the iteration, as well as the estimates of A and S (see Fig. ...