Sampling Theorems for Unsupervised Learning in Linear Inverse Problems

Julián Tachella, Dongdong Chen, Mike Davies

Research output: Working paperPreprint

Abstract / Description of output

Solving a linear inverse problem requires knowledge about the underlying signal model. In many applications, this model is a priori unknown and has to be learned from data. However, it is impossible to learn the model using observations obtained via a single incomplete measurement operator, as there is no information outside the range of the inverse operator, resulting in a chicken-and-egg problem: to learn the model we need reconstructed signals, but to reconstruct the signals we need to know the model. Two ways to overcome this limitation are using multiple measurement operators or assuming that the signal model is invariant to a certain group action. In this paper, we present necessary and sufficient sampling conditions for learning the signal model from partial measurements which only depend on the dimension of the model, and the number of operators or properties of the group action that the model is invariant to. As our results are agnostic of the learning algorithm, they shed light into the fundamental limitations of learning from incomplete data and have implications in a wide range set of practical algorithms, such as dictionary learning, matrix completion and deep neural networks.
Original languageEnglish
Publication statusPublished - 23 Mar 2022

Keywords / Materials (for Non-textual outputs)

  • stat.ML
  • cs.LG
  • eess.IV
  • 68U10
  • I.4.5; I.2.10; G.3

Fingerprint

Dive into the research topics of 'Sampling Theorems for Unsupervised Learning in Linear Inverse Problems'. Together they form a unique fingerprint.

Cite this