Recipes for Stable Linear Embeddings From Hilbert Spaces to {\mathbb {R}}^{m}

DOI10.1109/TIT.2017.2664858MaRDI QIDQ5280890zbMATH OpenOpenAlexFDO

Authors Gilles Puy, M. E. Davies, Rémi Gribonval

Publication date 27 July 2017

Published in IEEE Transactions on Information Theory (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1509.06947

Mathematics Subject Classification ID

Signal theory (characterization, reconstruction, filtering, etc.) (94A12)

Abstract: We consider the problem of constructing a linear map from a Hilbert space $mathcalH$ (possibly infinite dimensional) to $mathbb{R}^m$ that satisfies a restricted isometry property (RIP) on an arbitrary signal model $mathcalSsubsetmathcalH$. We present a generic framework that handles a large class of low-dimensional subsets but also unstructured and structured linear maps. We provide a simple recipe to prove that a random linear map satisfies a general RIP on $mathcalS$ with high probability. We also describe a generic technique to construct linear maps that satisfy the RIP. Finally, we detail how to use our results in several examples, which allow us to recover and extend many known compressive sampling results.

Cited in

(6)

• Generalizing CoSaMP to signals from a union of low dimensional linear subspaces
• Compressive statistical learning with random feature moments
• Time for dithering: fast and quantized random embeddings via the restricted isometry property
• On the Absence of Uniform Recovery in Many Real-World Applications of Compressed Sensing and the Restricted Isometry Property and Nullspace Property in Levels
• Sample complexity bounds for the local convergence of least squares approximation
• A theory of optimal convex regularization for low-dimensional recovery

This page was built for publication: Recipes for Stable Linear Embeddings From Hilbert Spaces to $ {\mathbb {R}}^{m}$

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5280890)