Research map / Learning theory and algorithms / Matrix and tensor methods

Sparse blind deconvolution via thresholded Wirtinger flow

NeurIPS 2026 · Learning theory and algorithms › Matrix and tensor methods · approach: Signal recovery and phase retrieval

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Studies sparse blind deconvolution (recovering an unknown kernel and sparse signal from their convolution) via a thresholded Wirtinger flow method.

Problem

Blind deconvolution is intrinsically ill-posed when both the kernel and the signal are unknown; meaningful recovery requires structural assumptions such as sparsity of the signal.

Approach

Applies thresholded Wirtinger flow, a nonconvex gradient-based method with thresholding that exploits sparsity of the signal, to recover the kernel and signal from their convolution. Details beyond the title and abstract are not available.

Experimental setup

Builds on: Wirtinger flow

Closest papers by meaning

• Structured Local Minima in Sparse Blind Deconvolution (NeurIPS 2018 · similarity 0.90)
• A Nonconvex Approach for Exact and Efficient Multichannel Sparse Blind Deconvolution (NeurIPS 2019 · similarity 0.89)
• Provable Non-convex Phase Retrieval with Outliers: Median TruncatedWirtinger Flow (ICML 2016 · similarity 0.87)
• Global Geometry of Multichannel Sparse Blind Deconvolution on the Sphere (NeurIPS 2018 · similarity 0.87)
• On the Global Geometry of Sphere-Constrained Sparse Blind Deconvolution (CVPR 2017 · similarity 0.86)
• The Price of Sparsity: Sufficient Conditions for Sparse Recovery using Sparse and Sparsified Measurements (NeurIPS 2025 · similarity 0.85)