Author: Ali Punjani; Haowei Zhang; David J. Fleet
Title: Non-uniform refinement: Adaptive regularization improves single particle cryo-EM reconstruction Document date: 2019_12_16
ID: bqwmx5dy_28
Snippet: Finally, the tuning parameters for non-uniform regularization are interpretable and relatively few in number. They include the order of the Butterworth kernel, the discretization of the parameter space, and the scalar relating Ï(x) and θ(x), called the adaptive window factor (AWF). In all experiments below we use a 2 nd -order Butterworth filter, and a fixed AWF parameter γ = 2. We discretize the regularization parameters into 50 possible valu.....
Document: Finally, the tuning parameters for non-uniform regularization are interpretable and relatively few in number. They include the order of the Butterworth kernel, the discretization of the parameter space, and the scalar relating Ï(x) and θ(x), called the adaptive window factor (AWF). In all experiments below we use a 2 nd -order Butterworth filter, and a fixed AWF parameter γ = 2. We discretize the regularization parameters into 50 possible values, equispaced in the Fourier domain to provide greater sensitivity to small scale changes at finer resolutions. We find that non-uniform refinement is approximately two times slower than uniform refinement in our current implementation.
Search related documents:
Co phrase search for related documents- adaptive window factor and non uniform refinement: 1
- approximately time and current implementation: 1
- AWF adaptive window factor and non uniform refinement: 1
- non uniform refinement and parameter space: 1
Co phrase search for related documents, hyperlinks ordered by date