Neurally Augmented ALISTA

by   Freya Behrens, et al.

It is well-established that many iterative sparse reconstruction algorithms can be unrolled to yield a learnable neural network for improved empirical performance. A prime example is learned ISTA (LISTA) where weights, step sizes and thresholds are learned from training data. Recently, Analytic LISTA (ALISTA) has been introduced, combining the strong empirical performance of a fully learned approach like LISTA, while retaining theoretical guarantees of classical compressed sensing algorithms and significantly reducing the number of parameters to learn. However, these parameters are trained to work in expectation, often leading to suboptimal reconstruction of individual targets. In this work we therefore introduce Neurally Augmented ALISTA, in which an LSTM network is used to compute step sizes and thresholds individually for each target vector during reconstruction. This adaptive approach is theoretically motivated by revisiting the recovery guarantees of ALISTA. We show that our approach further improves empirical performance in sparse reconstruction, in particular outperforming existing algorithms by an increasing margin as the compression ratio becomes more challenging.


page 1

page 2

page 3

page 4


Near-Optimal Adaptive Compressed Sensing

This paper proposes a simple adaptive sensing and group testing algorith...

Stabilizing Deep Tomographic Reconstruction Networks

While the field of deep tomographic reconstruction has been advancing ra...

Learned Experts' Assessment-based Reconstruction Network ("LEARN") for Sparse-data CT

Compressive sensing (CS) has proved effective for tomographic reconstruc...

Learned ISTA with Error-based Thresholding for Adaptive Sparse Coding

The learned iterative shrinkage thresholding algorithm (LISTA) introduce...

One Network to Solve All ROIs: Deep Learning CT for Any ROI using Differentiated Backprojection

Computed tomography for region-of-interest (ROI) reconstruction has adva...

Sparse Signal Recovery for Binary Compressed Sensing by Majority Voting Neural Networks

In this paper, we propose majority voting neural networks for sparse sig...