Difffit: Unlocking transferability of large diffusion models via simple parameter-efficient fine-tuning

E Xie, L Yao, H Shi, Z Liu, D Zhou… - Proceedings of the …, 2023 - openaccess.thecvf.com
Diffusion models have proven to be highly effective in generating high-quality images.
However, adapting large pre-trained diffusion models to new domains remains an open …

Theoretical perspectives on deep learning methods in inverse problems

J Scarlett, R Heckel, MRD Rodrigues… - IEEE journal on …, 2022 - ieeexplore.ieee.org
In recent years, there have been significant advances in the use of deep learning methods in
inverse problems such as denoising, compressive sensing, inpainting, and super-resolution …

Non-iterative recovery from nonlinear observations using generative models

J Liu, Z Liu - Proceedings of the IEEE/CVF Conference on …, 2022 - openaccess.thecvf.com
In this paper, we aim to estimate the direction of an underlying signal from its nonlinear
observations following the semi-parametric single index model (SIM). Unlike for …

Unsupervised deep learning for phase retrieval via teacher-student distillation

Y Quan, Z Chen, T Pang, H Ji - Proceedings of the AAAI Conference on …, 2023 - ojs.aaai.org
Phase retrieval (PR) is a challenging nonlinear inverse problem in scientific imaging that
involves reconstructing the phase of a signal from its intensity measurements. Recently …

Uniform Recovery Guarantees for Quantized Corrupted Sensing Using Structured or Generative Priors

J Chen, Z Liu, M Ding, MK Ng - SIAM Journal on Imaging Sciences, 2024 - SIAM
This paper studies quantized corrupted sensing where the measurements are contaminated
by unknown corruption and then quantized by a dithered uniform quantizer. We establish …

Misspecified phase retrieval with generative priors

Z Liu, X Wang, J Liu - Advances in Neural Information …, 2022 - proceedings.neurips.cc
In this paper, we study phase retrieval under model misspecification and generative priors.
In particular, we aim to estimate an $ n $-dimensional signal $\mathbf {x} $ from $ m $ iid …

Projected gradient descent algorithms for solving nonlinear inverse problems with generative priors

Z Liu, J Han - arXiv preprint arXiv:2209.10093, 2022 - arxiv.org
In this paper, we propose projected gradient descent (PGD) algorithms for signal estimation
from noisy nonlinear measurements. We assume that the unknown $ p $-dimensional signal …

Solving quadratic systems with full-rank matrices using sparse or generative priors

J Chen, MK Ng, Z Liu - arXiv preprint arXiv:2309.09032, 2023 - arxiv.org
The problem of recovering a signal $\boldsymbol x\in\mathbb {R}^ n $ from a quadratic
system $\{y_i=\boldsymbol x^\top\boldsymbol A_i\boldsymbol x,\i= 1,\ldots, m\} $ with full …

Efficient Algorithms for Non-gaussian Single Index Models with Generative Priors

J Chen, Z Liu - Proceedings of the AAAI Conference on Artificial …, 2024 - ojs.aaai.org
In this work, we focus on high-dimensional single index models with non-Gaussian sensing
vectors and generative priors. More specifically, our goal is to estimate the underlying signal …

Do algorithms and barriers for sparse principal component analysis extend to other structured settings?

G Wang, M Lou, A Pananjady - IEEE Transactions on Signal …, 2024 - ieeexplore.ieee.org
We study a principal component analysis problem under the spiked Wishart model in which
the structure in the signal is captured by a class of union-of-subspace models. This general …