implicitprior#
Module path: cuqi.implicitprior
Classes
Implicit Constrained GMRF (Gaussian Markov Random Field). |
|
Implicit Constrained Gaussian. |
|
This class defines (implicit) smoothed priors for which we can apply |
|
Implicit Nonnegative GMRF (Gaussian Markov Random Field). |
|
Implicit Nonnegative Gaussian. |
|
Implicit Regularized GMRF (Gaussian Markov Random Field). |
|
Implicit Regularized Gaussian. |
|
Implicit Regularized Unbounded Uniform. |
|
This class defines an implicit distribution associated with a restoration operator (eg denoiser). They are several works relating restorations operators with priors, see -Laumont et al. https://arxiv.org/abs/2103.04715 -Hu et al. https://openreview.net/pdf?id=x7d1qXEn1e We cannot sample from this distribution, neither compute its logpdf except in some cases. It allows us to apply algorithms such as MYULA and PnPULA. |
|
Alias for MoreauYoshidaPrior following Tweedie's formula framework. TweediePrior defines priors where gradients are computed based on Tweedie's identity that links MMSE (Minimum Mean Square Error) denoisers with the underlying smoothed prior, see: - Laumont et al. https://arxiv.org/abs/2103.04715 or https://doi.org/10.1137/21M1406349. |