Learned SVD: solving inverse problems via hybrid autoencoding

Yoeri E. Boink, Christoph Brune*

*Corresponding author for this work

Research output: Working paper

3 Downloads (Pure)

Abstract

Our world is full of physics-driven data where effective mappings between data manifolds are desired. There is an increasing demand for understanding combined model-based and data-driven methods. We propose a nonlinear, learned singular value decomposition (L-SVD), which combines autoencoders that simultaneously learn and connect latent codes for desired signals and given measurements. We provide a convergence analysis for a specifically structured L-SVD that acts as a regularisation method. In a more general setting, we investigate the topic of model reduction via data dimensionality reduction to obtain a regularised inversion. We present a promising direction for solving inverse problems in cases where the underlying physics are not fully understood or have very complex behaviour. We show that the building blocks of learned inversion maps can be obtained automatically, with improved performance upon classical methods and better interpretability than black-box methods.
Original languageEnglish
PublisherarXiv.org
Number of pages32
Publication statusSubmitted - 20 Dec 2020

Publication series

NameInverse Problems and Imaging
PublisherAmerican Institute of Mathematical Sciences
ISSN (Print)1930-8337

Keywords

  • inverse problems
  • neural networks
  • dimensionality reduction
  • autoencoders
  • SVD
  • regularisation

Fingerprint Dive into the research topics of 'Learned SVD: solving inverse problems via hybrid autoencoding'. Together they form a unique fingerprint.

Cite this