AI Paper Atlas

Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes

Variational Inference for Computational Imaging Inverse Problems

Francesco Tonolini, Jack Radford, Alex Turpin, Daniele Faccio, Roderick Murray-Smith

JMLR 2020 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Variational Bayesian deep learning · 31 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
train with proposed multi-fidelity variational framework instead of paired examples onlyCVAE · CelebA K=1,000 x2 down-samplingtest ELBO: -19827 -> 14553Improved
train with proposed framework instead of paired examples onlyCVAE · CelebA K=10,000 x2 down-samplingtest ELBO: 13450 -> 14763Improved
train with proposed framework instead of simulations onlyCVAE · CelebA K=1,000 x2 down-samplingtest ELBO: -208 -> 14553Improved
train with proposed framework instead of naive paired+simulations poolingCVAE · CelebA K=1,000 x2 down-samplingtest ELBO: 11300 -> 14553Improved
train with proposed framework instead of naive poolingCVAE · CIFAR10 K=10,000 x2 down-samplingtest ELBO: 12096 -> 14415Improved
increase paired examples K from 1,000 to 10,000 for standard CVAE (ablation)CVAE paired only · CelebA blurring sigma=1.5pxtest ELBO: -13872 -> 13502Improved
increase paired examples K from 1,000 to 10,000 for proposed framework (ablation)proposed framework · CelebA partial occlusiontest ELBO: 15134 -> 15187No clear effect
add unpaired targets and low-fidelity model (proposed) for scarce paired dataCVAE · CelebA 64x64 deblurring, K<few thousandPSNR and ELBOImproved
use more accurate observation model (ablation)proposed framework · CelebA blurring, low KPSNRImproved
use proposed variational framework instead of HIO and deep ANNholographic reconstruction · MNIST Fourier intensity (9,600 pairs)reconstruction qualityImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy