D
David L. Donoho
Researcher at Stanford University
Publications - 273
Citations - 115802
David L. Donoho is an academic researcher from Stanford University. The author has contributed to research in topics: Wavelet & Compressed sensing. The author has an hindex of 110, co-authored 271 publications receiving 108027 citations. Previous affiliations of David L. Donoho include University of California, Berkeley & Western Geophysical.
Papers
More filters
Posted Content
Optimal Shrinkage of Singular Values
Matan Gavish,David L. Donoho +1 more
TL;DR: In this article, the authors consider recovery of low-rank matrices from noisy data by shrinkage of singular values, in which a single, univariate nonlinearity is applied to each of the empirical singular values.
Journal ArticleDOI
Minimax risk of matrix denoising by singular value thresholding
David L. Donoho,Matan Gavish +1 more
TL;DR: In this article, the problem of matrix denoising is solved by applying soft thresholding to the singular values of the noisy measurement, where the noise matrix has i.i.d. Gaussian entries.
Posted Content
Microlocal Analysis of the Geometric Separation Problem
David L. Donoho,Gitta Kutyniok +1 more
TL;DR: A theoretical analysis is presented showing that accurate geometric separation of point and curve singularities can be achieved by minimizing the $\ell_1$ norm of the representing coefficients in two geometrically complementary frames: wavelets and curvelets.
Proceedings ArticleDOI
Breakdown Point of Model Selection When the Number of Variables Exceeds the Number of Observations
David L. Donoho,Victoria Stodden +1 more
TL;DR: This work points out that when p > n, there is a breakdown point for standard model selection schemes, such that model selection only works well below a certain critical complexity level depending on n/p, and applies this notion to some model selection algorithms (Forward Stepwise, LASSO, LARS) in the case where pGtn.
Faster Imaging with Randomly Perturbed, Undersampled Spirals and |L|_1 Reconstruction
TL;DR: A fast imaging method based on undersampled k-space spiral sampling and non-linear reconstruction, inspired by theoretical results in sparse signal recovery, allowing 50% undersampling by adapting spiral MR imaging and introducing randomness by perturbing the authors' spiral trajectories.