Scientific Organizers
-
Per Christian Hansen, Technical University of Denmark
-
Marcelo Pereyra, Heriot Watt University
-
Yiqiu Dong, Technical University of Denmark
About:
Inverse problems and imaging applications are everywhere, and often there is a need for assessing the uncertainty of the reconstructions due to noise, model errors, etc. by means of uncertainty quantification (UQ). Development of UQ methods requires collaboration across several areas.
This workshop will bring together specialists in UQ for inverse problems and imaging, and we invite talks related to the development of theory, methodology, and software. We also invite talks about interesting applications of UQ in imaging. The goal is to stimulate networking and collaboration between researchers and students in these areas, and to present state-of-the-art research results.
The workshop is supported by the project CUQI (Computational Uncertainty Quantification for Inverse problems) funded by the Villum Foundation (grant no. 25893) and by UKRI EPSRC BLOOM (EP/V006134/1) and LEXCI (EP/W007673/1).
Plenary Speakers
- Yoann Altmann, Heriot-Watt University
- Tatiana Bubba, University of Bath
- Per Christian Hansen, Technical Univ. of Denmark
- Aku Seppänen, University of Eastern Finland
- Julian Tachella , CNRS and ENS de Lyon
- Faouzi Triki, Grenoble-Alpes University
CUQIpy Software Tranining Course
CUQIpy is a python software package for computational uncertainty quantification for inverse problems, developed in the CUQI research project.
Before the main workshop, we give a training course on this software. Participants will learn to use CUQIpy to model statistical inverse problems and perform UQ on them. The course includes hands-on tutorials (bring your laptop!) with examples from image deblurring, X-ray CT, and inverse problems based on partial differential equations. Half of the course is devoted to working on a small use-case with CUQIpy, and participants are encouraged to bring their own case and data.
Participation
Participation in person only.
Programme
Monday morning is devoted to a brief tutorial on Bayesian inference and UQ for inverse problems.
The CUQIpy training course lasts from Monday noon until Tuesday noon; Monday evening is available for the nerds.
The core workshop lasts from Tuesday noon until Friday noon, and consists of plenary talks, contributed talks, and poster sessions. There will be a welcome reception on Tuesday evening, and on Thursday evening there will be a guided tour followed by the workshop dinner.
For those who stay on Friday afternoon, we arrange a social event – perhaps a visit to a whisky or gin distillery.
Scam Warning
We are aware of a number of current scams targeting participants at ICMS workshops in relation to registration or accommodation bookings. If you are approached by a third party (eg travellerpoint.org, Conference Committee or Royal Visit) asking for booking or payment details, please ignore. Please remember:
i) ICMS never uses third parties to do our administration for events: messages will come directly from ICMS staff
ii) ICMS will never ask participants for credit card or bank details
iii) If you have any doubt about an email you receive please get in touch
Programme
MONDAY 16 September 2024 | ||
CUQIpy training course registration with refreshments | ||
Welcome and Housekeeping | ||
Marcelo Pereyra, Heriot Watt University | Tutorial | |
Lunch | ||
CUQIpy | ||
Pizza night at Bayes Centre | ||
CUQIpy hackathon | ||
TUESDAY 17 September 2024 | ||
CUQIpy | ||
Lunch & Core workshop registration | ||
Welcome and Housekeeping | ||
Per Chr. Hansen, Technical University of Denmark | Plenary talk: CUQI – A Research Initiative in Uncertainty Quantification for Inverse Problems | |
Jasper M. Everink, Technical University of Denmark (DTU) | Contributed talk: Sparsity-promoting implicit priors using regularized Gaussian distributions | |
Amal Alghamdi, Technical University of Denmark | Contributed talk: Characterizing transport model in cochlear aqueduct using CT data and a Bayesian inversion approach | |
Poster pitches | ||
5 posters & Refreshments | Posters: Maud Biquard, Jonas Bresch, Jonathan Spence, Vincent Verhoeven, Jakob S. Jørgensen | |
Welcome reception at Bayes Centre | ||
WEDNESDAY 18 September 2024 | ||
Aku Seppänen, University of Eastern Finland | Plenary talk: Non-linear difference imaging and tolerance to modelling errors in electrical impedance tomography | |
Richard Huber, Laboratoire TIMC, Université Grenoble Alpes, CNRS | Contributed talk: Approximation Errors of Discretizations of the Radon Transform | |
Refreshments | ||
Andrea Sebastiani, University of Modena and Reggio Emilia | Contributed talk: Improving Plug-and-Play convergence with stepsize selection | |
Rémi Laumon, Technical University of Denmark | Contributed talk: A Comparative Analysis of RTO and Langevin Sampling Methods in Bayesian Inverse Problems | |
Lunch | ||
Tatiana Bubba, University of Bath | Plenary talk: Proximal operators learning meets unrolling in imaging inverse problems | |
Charlesquin Mbakam, Heriot-Watt university | Contributed talk: Empirical Bayesian PnP imaging with denoising diffusion priors | |
Poster pitches | ||
5 posters & Refreshments | Posters: Margaret Duff, Abderrahim Halimi, Marien RENAUD, Xuebin Zhao | |
THURSDAY 19 September 2024 | ||
Yoann Altman, Heriot-Watt University | Plenary talk: Expectation-Propagation for large-scale imaging and computer vision problems | |
Barbara Pascal, Nantes Université, École Centrale Nantes | Contributed talk: Proximal-Langevin samplers for nonsmooth composite posteriors: Application to the estimation of Covid19 reproduction number | |
Refreshments | ||
Elisa Crabu, University of Cagliari | Contributed talk: Dealing with non-Lambertianity in Photometric Stereo | |
Joshua Hellier, University of Manchester | Contributed talk: Target Localisation in Multistatic Pulse-Doppler Radar | |
Lunch | ||
Faouzi Triki, Grenoble-Alpes University | Plenary talk: PASSIVE MEASUREMENTS∗ Stability for the inverse source problem with passive measurements | |
Dawit Abiy Hailu, Helmholtz-zentrum Hereon | Contributed talk: Generative models for solving Inverse problems | |
Poster pitches | ||
5 posters & Refreshments | Posters: Lara Baalbaki, Emma Hart, Nathanaël Munier, Daniel Todt, Tania Andreea Goia & Naoki Sakai (joint poster) | |
Guided city tour | ||
Workshop Dinner, hosted at Apex Grassmarket Hotel | ||
FRIDAY 20 September 2024 | ||
Julian Tachella, CNRS and ENS de Lyon | Plenary talk: UNSURE: Unknown Noise level Stein's Unbiased Risk Estimator | |
Teresa Klatzer, University of Edinburgh | Contributed talk: Bayesian Computation with Plug and Play Priors for Poisson Inverse Problems | |
Poster pitches | ||
5 posters & Refreshments & Lunch | Posters: Hubert Leterme, Savvas Melidonis, Giacomo Varini, Yiyao Zhang, Yiming Xi | |
Distillery tour |