Generative Inference and Calibration


Generative models are experiencing a second youth in imaging and scientific inference. New ideas include injective models for sampling high-dimensional posteriors, theoretical advances on statistical, approximation-theoretic, and topological questions, generating continuous functions which dovetail with the downstream PDE solvers, and creative uses of generative models to probe performance limits of inference systems. The session “Generative Inference and Calibration” brings together prominent researchers spearheading these exciting new directions.

Organisers: Ivan Dokmanić (Chair) & Philip Schniter (Committee member)

Session Schedule

08:00 - 08:20 Invited talk -
Gen-Alpha Generative Models for Imaging
Ivan Dokmanić
08:20 - 08:40 Invited talk -
Continuous Generative Neural Networks
Giovanni Alberti
08:40 - 09:00 Invited talk -
Deep and Shallow Generative Models of Images
Yair Weiss
09:00 - 09:20 Invited talk -
Learning to Bound: A Generative Cram ́er-Rao Bound
Yoram Bresler
09:20 - 09:40 Invited talk -
Interferometric Phase Image Estimation Using Importance Sampling
Mario Figueiredo
09:40 - 11:00 Invited poster -
Deep Invertible Approximation of Topologically Rich Maps between Manifolds
Maarten de Hoop
09:40 - 11:00 Contributed poster -
Validation Diagnostics for SBI algorithms based on Normalizing Flows
Julia Linhart*
09:40 - 11:00 Contributed poster -
Stable deep MRI reconstruction using Generative Priors
Martin Zach*
09:40 - 11:00 Committee member poster -
A Regularized Conditional GAN for Posterior Sampling in Inverse Problems
Phil Schniter