Applications of deep learning in Bayesian inversion

Add to your list(s) Download to your calendar using vCal

• Ozan Öktem (KTH and Alan Turing Institute)
• Tuesday 20 November 2018, 13:00-14:00
• MR5, Centre for Mathematical Sciences.

If you have a question about this talk, please contact Carola-Bibiane Schoenlieb.

This talk has been canceled/deleted

The talk will show how deep neural networks can be used to compute a Bayes estimator in a computationally feasible manner without explicitly specifying a prior or probability of data. The prior and probability of data are implicitly contained in supervised data that is used to train the deep neural network, whereas the data likelihood is explicitly included into the network architecture. Next, we also show how to use generative adversarial networks to sample from the posterior in a computationally feasible manner. Both these approaches are generic, and their performance is demonstrated for tomographic reconstruction in a clinical setting.

This talk is part of the Applied and Computational Analysis series.

This talk is included in these lists:

This talk is not included in any other list

Note that ex-directory lists are not shown.