Quantized Compressed Sensing with Score-Based Generative Models

  title={Quantized Compressed Sensing with Score-Based Generative Models},
  author={Xiangming Meng and Yoshiyuki Kabashima},
We consider the general problem of recovering a high-dimensional signal from noisy quantized measurements. Quantization, especially coarse quantization such as 1-bit sign measurements, leads to severe information loss and thus a good prior knowledge of the unknown signal is helpful for accurate recovery. Motivated by the power of score-based generative models (SGM, also known as diffusion models) in capturing the rich structure of natural signals beyond simple sparsity, we propose an… 

Diffusion Model Based Posterior Sampling for Noisy Linear Inverse Problems

An unsupervised general-purpose sampling approach called diffusion model based posterior sampling (DMPS) to reconstruct the unknown signal from noisy linear measurements and achieves highly competitive or even better performances on various tasks while being 3 times faster than the leading competitor.

QCM-SGM+: Improved Quantized Compressed Sensing With Score-Based Generative Models for General Sensing Matrices

An improved version of QCS-SGM, which is based on a Bayesian inference perspective of the likelihood score computation, whereby an expectation propagation algorithm is proposed to approximately compute the likelihood scores, which works well for general matrices.



Sample Complexity Bounds for 1-bit Compressive Sensing and Binary Stable Embeddings with Generative Priors

It is demonstrated that the Binary $\epsilon$-Stable Embedding property, which characterizes the robustness of the reconstruction to measurement errors and noise, also holds for 1-bit compressive sensing with Lipschitz continuous generative models with sufficiently many Gaussian measurements.

Non-Iterative Recovery from Nonlinear Observations using Generative Models

  • Jiulong LiuZhaoqiang Liu
  • Computer Science
    2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2022
This paper aims to estimate the direction of an underlying signal from its nonlinear observations following the semi-parametric single index model (SIM), and shows that the non-iterative method significantly outperforms a state-of-the-art iterative method in terms of both accuracy and efficiency.

Compressed Sensing using Generative Models

This work shows how to achieve guarantees similar to standard compressed sensing but without employing sparsity at all, and proves that, if G is L-Lipschitz, then roughly O(k log L) random Gaussian measurements suffice for an l2/l2 recovery guarantee.

Instance-Optimal Compressed Sensing via Posterior Sampling

The posterior sampling estimator for deep generative priors is implemented using Langevin dynamics, and it is empirically found that it produces accurate estimates with more diversity than MAP.

Dequantizing Compressed Sensing: When Oversampling and Non-Gaussian Constraints Combine

A new class of convex optimization programs, or decoders, coined Basis Pursuit DeQuantizer of moment p (BPDQp), that model the quantization distortion more faithfully than the commonly used Basis pursuit DeNoise (B PDN) program are presented.

Deep Compressed Sensing

Borrowing insights from the CS perspective, a novel way of improving GANs using gradient information from the discriminator is developed and it is shown that Generative Adversarial Nets (GANs) can be viewed as a special case in this family of models.

1-Bit compressive sensing

This paper reformulates the problem by treating the 1-bit measurements as sign constraints and further constraining the optimization to recover a signal on the unit sphere, and demonstrates that this approach performs significantly better compared to the classical compressive sensing reconstruction methods, even as the signal becomes less sparse and as the number of measurements increases.

Robust 1-Bit Compressive Sensing via Binary Stable Embeddings of Sparse Vectors

This paper investigates an alternative CS approach that shifts the emphasis from the sampling rate to the number of bits per measurement, and introduces the binary iterative hard thresholding algorithm for signal reconstruction from 1-bit measurements that offers state-of-the-art performance.

Learning and 1-bit Compressed Sensing under Asymmetric Noise

Algorithms with nearly optimal guarantees for both problems under two realistic noise models, bounded (Massart) noise and adversarial (agnostic) noise, when the measurements x_i’s are drawn from any isotropic log-concave distribution are introduced.

Statistical mechanics approach to 1-bit compressed sensing

It is shown that the signal recovery performance predicted by the replica method under the replica symmetric ansatz is in good consistency with experimental results of an approximate recovery algorithm developed earlier, suggesting that the l1-based recovery problem typically has many local optima of a similar recovery accuracy.