Variational Free Energies for Compressed Sensing

Abstract

We consider the variational free energy approach for compressed sensing. We first show that the naïve mean field approach performs remarkably well when coupled with a noise learning procedure. We also notice that it leads to the same equations as those used for iterative thresholding. We then discuss the Bethe free energy and how it corresponds to the fixed points of the approximate message passing algorithm. In both cases, we test numerically the direct optimization of the free energies as a converging sparse-estimationalgorithm.

5 pages, 3 figures

References in corpus (1)

• Probabilistic Reconstruction in Compressed Sensing: Algorithms, Phase Diagrams, and Threshold Achieving Matrices

Cited by in corpus (8)

• Compressive Phase Retrieval via Generalized Approximate Message Passing
• Adaptive Damping and Mean Removal for the Generalized Approximate Message Passing Algorithm
• A Survey of Stochastic Simulation and Optimization Methods in Signal Processing
• Approximate message-passing decoder and capacity-achieving sparse superposition codes
• Phase transitions and sample complexity in Bayes-optimal matrix factorization
• High-temperature Expansions and Message Passing Algorithms
• Phase recovery from a Bayesian point of view: the variational approach
• Streaming Bayesian inference: theoretical limits and mini-batch approximate message-passing