Fundamental performance limits for ideal decoders in high-dimensional linear inverse problems

Abstract

This paper focuses on characterizing the fundamental performance limits that can be expected from an ideal decoder given a general model, ie, a general subset of "simple" vectors of interest. First, we extend the so-called notion of instance optimality of a decoder to settings where one only wishes to reconstruct some part of the original high dimensional vector from a low-dimensional observation. This covers practical settings such as medical imaging of a region of interest, or audio source separation when one is only interested in estimating the contribution of a specific instrument to a musical recording. We define instance optimality relatively to a model much beyond the traditional framework of sparse recovery, and characterize the existence of an instance optimal decoder in terms of joint properties of the model and the considered linear operator. Noiseless and noise-robust settings are both considered. We show somewhat surprisingly that the existence of noise-aware instance optimal decoders for all noise levels implies the existence of a noise-blind decoder. A consequence of our results is that for models that are rich enough to contain an orthonormal basis, the existence of an L2/L2 instance optimal decoder is only possible when the linear operator is not substantially dimension-reducing. This covers well-known cases (sparse vectors, low-rank matrices) as well as a number of seemingly new situations (structured sparsity and sparse inverse covariance matrices for instance). We exhibit an operator-dependent norm which, under a model-specific generalization of the Restricted Isometry Property (RIP), always yields a feasible instance optimality property. This norm can be upper bounded by an atomic norm relative to the considered model.

To appear in IEEE Transactions on Information Theory

References in corpus (1)

• New Analysis of Manifold Embeddings and Signal Recovery from Compressive Measurements

Cited by in corpus (13)

• The basins of attraction of the global minimizers of the non-convex sparse spike estimation problem
• On the absence of the RIP in real-world applications of compressed sensing and the RIP in levels
• The basins of attraction of the global minimizers of non-convex inverse problems with low-dimensional models in infinite dimension
• Beyond Moore-Penrose Part I: Generalized Inverses that Minimize Matrix Norms
• Compressed sensing with local structure: uniform recovery guarantees for the sparsity in levels class
• Recipes for stable linear embeddings from Hilbert spaces to R^m
• Instance Optimal Decoding and the Restricted Isometry Property
• Infinite-dimensional compressed sensing and function interpolation
• Unsupervised Learning From Incomplete Measurements for Inverse Problems
• Optimality of 1-norm regularization among weighted 1-norms for sparse recovery: a case study on how to find optimal regularizations
• Instance optimality in phase retrieval
• Is the 1-norm the best convex sparse regularization?
• Fast sensor placement by enlarging principle submatrix for large-scale linear inverse problems