MinVariance

class colibri.regularizers.MinVariance(param=0.01)

Bases: Module

Minimum Variance Regularization.

Code adapted from [2] Jacome, Roman, Pablo Gomez, and Henry Arguello. “Middle output regularized end-to-end optimization for computational imaging.” Optica 10.11 (2023): 1421-1431.

\[\begin{equation*} R(\mathbf{y}) = \mu\left\|\sigma_{\mathbf{y}}\right\|_2 \end{equation*}\]

where \(\sigma_{\mathbf{y}}\) is the standard deviation of the input tensor \(\mathbf{y}\in\yset\).

Parameters:

param (float) – Regularization parameter.

forward(y)

Compute minimum variance regularization term.

Parameters:

y (torch.Tensor) – Input tensor.

Returns:

Minimum variance regularization term.

Return type:

torch.Tensor

Examples using MinVariance:

Demo Colibri.

Demo Colibri.