Search Results for author: Oscar Leong

Found 12 papers, 2 papers with code

The Star Geometry of Critic-Based Regularizer Learning

no code implementations29 Aug 2024 Oscar Leong, Eliza O'Reilly, Yong Sheng Soh

Variational regularization is a classical technique to solve statistical inference tasks and inverse problems, with modern data-driven approaches parameterizing regularizers via deep neural networks showcasing impressive empirical performance.

Flow Priors for Linear Inverse Problems via Iterative Corrupted Trajectory Matching

2 code implementations29 May 2024 Yasi Zhang, Peiyu Yu, Yaxuan Zhu, Yingshan Chang, Feng Gao, Ying Nian Wu, Oscar Leong

We validate our approach for various linear inverse problems, such as super-resolution, deblurring, inpainting, and compressed sensing, and demonstrate that we can outperform other methods based on flow matching.

Deblurring Image Generation +1

Score-Based Diffusion Models for Photoacoustic Tomography Image Reconstruction

no code implementations30 Mar 2024 Sreemanti Dey, Snigdha Saha, Berthy T. Feng, Manxiu Cui, Laure Delisle, Oscar Leong, Lihong V. Wang, Katherine L. Bouman

Photoacoustic tomography (PAT) is a rapidly-evolving medical imaging modality that combines optical absorption contrast with ultrasound imaging depth.

Image Reconstruction

Discovering Structure From Corruption for Unsupervised Image Reconstruction

no code implementations12 Apr 2023 Oscar Leong, Angela F. Gao, He Sun, Katherine L. Bouman

We show that such a set of inverse problems can be solved simultaneously without the use of a spatial image prior by instead inferring a shared image generator with a low-dimensional latent space.

Denoising Image Reconstruction +1

Image Reconstruction without Explicit Priors

no code implementations21 Mar 2023 Angela F. Gao, Oscar Leong, He Sun, Katherine L. Bouman

We show that such a set of inverse problems can be solved simultaneously by learning a shared image generator with a low-dimensional latent space.

Image Reconstruction

Optimal Regularization for a Data Source

no code implementations27 Dec 2022 Oscar Leong, Eliza O'Reilly, Yong Sheng Soh, Venkat Chandrasekaran

In this paper, we seek a systematic understanding of the power and the limitations of convex regularization by investigating the following questions: Given a distribution, what is the optimal regularizer for data drawn from the distribution?

Dictionary Learning

Alternating Phase Langevin Sampling with Implicit Denoiser Priors for Phase Retrieval

no code implementations2 Nov 2022 Rohun Agrawal, Oscar Leong

Phase retrieval is the nonlinear inverse problem of recovering a true signal from its Fourier magnitude measurements.

Denoising Retrieval

Optimal Sample Complexity of Subgradient Descent for Amplitude Flow via Non-Lipschitz Matrix Concentration

no code implementations31 Oct 2020 Paul Hand, Oscar Leong, Vladislav Voroninski

We establish local convergence of subgradient descent with optimal sample complexity based on the uniform concentration of a random, discontinuous matrix-valued operator arising from the objective's gradient dynamics.

Compressive Phase Retrieval: Optimal Sample Complexity with Deep Generative Priors

no code implementations24 Aug 2020 Paul Hand, Oscar Leong, Vladislav Voroninski

Advances in compressive sensing provided reconstruction algorithms of sparse signals from linear measurements with optimal sample complexity, but natural extensions of this methodology to nonlinear inverse problems have been met with potentially fundamental sample complexity bottlenecks.

Compressive Sensing Retrieval

Low Shot Learning with Untrained Neural Networks for Imaging Inverse Problems

no code implementations NeurIPS Workshop Deep_Invers 2019 Oscar Leong, Wesam Sakla

In particular, how can one use the availability of a small amount of data (even $5-25$ examples) to one's advantage in solving these inverse problems and can a system's performance increase as the amount of data increases as well?

Colorization

Invertible generative models for inverse problems: mitigating representation error and dataset bias

1 code implementation28 May 2019 Muhammad Asim, Max Daniels, Oscar Leong, Ali Ahmed, Paul Hand

For compressive sensing, invertible priors can yield higher accuracy than sparsity priors across almost all undersampling ratios, and due to their lack of representation error, invertible priors can yield better reconstructions than GAN priors for images that have rare features of variation within the biased training set, including out-of-distribution natural images.

Compressive Sensing Decoder +2

Phase Retrieval Under a Generative Prior

no code implementations NeurIPS 2018 Paul Hand, Oscar Leong, Vladislav Voroninski

Our formulation has provably favorable global geometry for gradient methods, as soon as $m = O(kd^2\log n)$, where $d$ is the depth of the network.

Retrieval

Cannot find the paper you are looking for? You can Submit a new open access paper.