# Generative Model Adversarial Training for Deep Compressed Sensing

arXiv: https://arxiv.org/abs/2106.10696 · HTML: https://mlsanity.com/paper/2106.10696/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 14 of its 26 references that are in the map.
- First posted on arXiv 20 Jun 2021.

## Builds on (most cited first, 14 of 14)

- [Universal adversarial perturbations](https://mlsanity.com/paper/1610.08401/) (2016)
- [Compressed Sensing using Generative Models](https://mlsanity.com/paper/1703.03208/) (2017)
- [Generative Adversarial Network in Medical Imaging: A Review](https://mlsanity.com/paper/1809.07294/) (2018)
- [Compressed Sensing MRI Reconstruction using a Generative Adversarial Network with a Cyclic Loss](https://mlsanity.com/paper/1709.00753/) (2017)
- [Defense against Universal Adversarial Perturbations](https://mlsanity.com/paper/1711.05929/) (2017)
- [Compressed Sensing with Deep Image Prior and Learned Regularization](https://mlsanity.com/paper/1806.06438/) (2018)
- [Deep Compressed Sensing](https://mlsanity.com/paper/1905.06723/) (2019)
- [The Robust Manifold Defense: Adversarial Training using Generative Models](https://mlsanity.com/paper/1712.09196/) (2017)
- [RecurJac: An Efficient Recursive Algorithm for Bounding Jacobian Matrix of Neural Networks and Its Applications](https://mlsanity.com/paper/1810.11783/) (2018)
- [Deep Generative Adversarial Networks for Compressed Sensing Automates MRI](https://mlsanity.com/paper/1706.00051/) (2017)
- [SEGAN: Structure-Enhanced Generative Adversarial Network for Compressed Sensing MRI Reconstruction](https://mlsanity.com/paper/1902.06455/) (2019)
- [A Novel Approach to Quantized Matrix Completion Using Huber Loss Measure](https://mlsanity.com/paper/1810.12460/) (2018)
- [Transduction with Matrix Completion Using Smoothed Rank Function](https://mlsanity.com/paper/1805.07561/) (2018)
- [Iterative Null-space Projection Method with Adaptive Thresholding in Sparse Signal Recovery and Matrix Completion](https://mlsanity.com/paper/1610.00287/) (2016)

## Abstract (arXiv)

Deep compressed sensing assumes the data has sparse representation in a latent space, i.e., it is intrinsically of low-dimension. The original data is assumed to be mapped from a low-dimensional space through a low-to-high-dimensional generator. In this work, we propound how to design such a low-to-high dimensional deep learning-based generator suiting for compressed sensing, while satisfying robustness to universal adversarial perturbations in the latent domain. We also justify why the noise is considered in the latent space. The work is also buttressed with theoretical analysis on the robustness of the trained generator to adversarial perturbations. Experiments on real-world datasets are provided to substantiate the efficacy of the proposed generative model adversarial training for deep compressed sensing.