Academy of Mathematics and Systems Science, CAS Colloquia & Seminars
Speaker:
Dr. Jonas Adler, Department of Mathematics, Royal Institute of Technology (KTH)
Inviter:
陈冲
Title:
Deep Learning for Image Reconstruction
Time & Venue:
2018.10.19 15:00-16:00 Z311
Abstract:
Deep learning has shown tremendous success in solving various tasks in several fields of science such as in image and natural language processing. Recently, it has also been applied to solve inverse problems and empirical evidence in image reconstruction points to exponential improvements in both performance and run-time over classical approaches. In this talk we outline some of these recent developments. In particular, we'll introduce "Learned Iterative Reconstruction", a method which relies on a fusion of model-anddata-driven approaches for solving inverse problems. We'll also show how these methods can be extended to "Deep Bayesian Inversion", a family of methods that allows us to perform uncertainty quantification using deep learning. CV: Jonas Adler holds a MSc in Engineering Physics. Since 2013, he is a Research Scientist with Elekta and since 2015 he is also pursuing a PhD in applied mathematics at KTH - Royal Institute of Technology under the supervision of Prof. Ozan ?ktem. His research is on inverse problems, machine learning and their intersection.He is also an active open source developer, contributing to several packages.