M22: Inverse Problems with Data-Driven Methods and Deep Learning

We are currently experiencing a paradigm shift in image reconstruction. Robust mathematical inversion algorithms are combined with emerging methods in data science to achieve state-of-the-art results in a wide range of inverse problems. A successful application of these methods in practice involves a thorough understanding of their mathematical properties and the underlying imaging physics. This minisymposium aims to bring experts in data-driven methods and deep learning for inverse problems together and provides an overview of learned image reconstruction approaches, mathematical insights, and real-world applications.

Organizers:
Tatiana Bubba, University of Bath, UK, This email address is being protected from spambots. You need JavaScript enabled to view it.
Martin Genzel, Utrecht University, Netherlands, This email address is being protected from spambots. You need JavaScript enabled to view it.
Andreas Hauptmann, University of Oulu, Finland, This email address is being protected from spambots. You need JavaScript enabled to view it.
Maximilian März, Technical University of Berlin, Germany, This email address is being protected from spambots. You need JavaScript enabled to view it.

Invited Speakers :
Vegard Antun, University of Oslo, Norway, This email address is being protected from spambots. You need JavaScript enabled to view it.
On hallucinations and barriers in deep learning for linear inverse problems

Simon Arridge, University College London, UK, This email address is being protected from spambots. You need JavaScript enabled to view it.
Learned approximations of nonlinear operators for inverse problems

Angelica Aviles-Rivero, University of Cambridge, UK, This email address is being protected from spambots. You need JavaScript enabled to view it.
Energy Models for Better Pseudo-Labels: Semi-Supervised Classification with the 1-Laplacian Graph Energy

Riccardo Barbanos, University College London, UK, This email address is being protected from spambots. You need JavaScript enabled to view it.
Understanding Pretraining in Deep Image Prior

Sören Dittmer, University of Cambridge, UK, This email address is being protected from spambots. You need JavaScript enabled to view it.
Ground truth free denoising by optimal transport

Margaret Duff, University of Bath, UK, This email address is being protected from spambots. You need JavaScript enabled to view it.
Regularising Inverse Problems with Generative Machine Learning Models

Reinhard Heckel, TU Munich, Germany, This email address is being protected from spambots. You need JavaScript enabled to view it.
TBA

Johannes Hertrich, TU Berlin, Germany, This email address is being protected from spambots. You need JavaScript enabled to view it.
Stochastic normalizing flows for inverse problems: a Markov chains viewpoint

Peter Jung, TU Berlin, Germany, This email address is being protected from spambots. You need JavaScript enabled to view it.
Towards Neurally-Augmented Algorithms for Solving Inverse Problems

Ulugbek Kamilov, Washington University, USA, This email address is being protected from spambots. You need JavaScript enabled to view it.
Recovery Analysis for Plug-and-Play Priors using the Restricted Eigenvalue Condition

Maximilian Kiss, CWI, Netherlands, This email address is being protected from spambots. You need JavaScript enabled to view it.
2DeteCT: A large 2D expandable, trainable, experimental computed tomography data collection for machine learning

Jan Macdonald, TU Berlin, Germany, This email address is being protected from spambots. You need JavaScript enabled to view it.
Solving Inverse Problems With Deep Neural Networks - Robustness Included?

Nicole Mücke, TU Braunschweig, Germany, This email address is being protected from spambots. You need JavaScript enabled to view it.
SGD for solving non-linear inverse Problems with application to Neural Network learning

Subhadip Mukherjee, University of Cambridge, UK, This email address is being protected from spambots. You need JavaScript enabled to view it.
Data-driven adversarial regularization for imaging inverse problems

Jenni Poimala, University of Oulu, Finland, This email address is being protected from spambots. You need JavaScript enabled to view it.
Learned speed of sound correction for photoacoustic tomography

Tom Tirer, Tel-Aviv University, Israel, This email address is being protected from spambots. You need JavaScript enabled to view it.
Solving Ill-Posed Inverse Problems with Pretrained Denoisers, GANs and Super-Resolvers: The BP Term and the Correction Filter

Pierre Weiss, CNRS, France, This email address is being protected from spambots. You need JavaScript enabled to view it.
Deep learning based identification of operators in blind inverse problems

Samy Wu Fung, Colorado School of Mines, USA, This email address is being protected from spambots. You need JavaScript enabled to view it.
Efficient Training of Infinite-Depth Neural Networks via Jacobian-Free Backpropagation

Jong Chul Ye, KAIST, South Korea, This email address is being protected from spambots. You need JavaScript enabled to view it.
TBA