Research map / Learning theory and algorithms / Matrix and tensor methods
Short and Sparse Deconvolution --- A Geometric Approach
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use data-driven initialization instead of uniform random initialization | ADM and iADM · Synthetic SaSD, coherent and incoherent kernels | convergence speed and solution quality | Helped |
| Add momentum acceleration | ADM for synthetic SaSD · Synthetic SaSD, especially coherent kernels | convergence rate | Helped |
| Use homotopy continuation with a gradually decreasing sparsity penalty | iADM for synthetic SaSD · Synthetic SaSD, coherent and incoherent kernels | convergence rate | Helped |
| Replace iADM with ADMM | Bilinear Lasso optimization · Synthetic SaSD with incoherent kernels | convergence | Helped |
| Replace iADM with ADMM | Bilinear Lasso optimization · Synthetic SaSD with coherent kernels | reliability | Hurt |
| Use iADM with homotopy instead of the compared optimization methods | Bilinear Lasso optimization · Synthetic SaSD, coherent and incoherent kernels | convergence | Helped |
| Use nonnegative constraints on the activation map | SaSD · Synthetic SaSD with nonnegative activations | recoverable activation density | Helped |
| Add reweighting to iADM | iADM · Simulated calcium imaging, AR(1) and AR(2) | kernel estimation accuracy | Helped |
| Add reweighting to iADM | iADM · Simulated calcium imaging with AR(2) dynamics | spike-time resolution | Helped |
| Add reweighting to iADM | iADM · Real calcium imaging | spike-train estimation | Helped |
| Apply the method to similar rather than distinct neuron waveforms | Reweighted iADM for spike sorting · Difficult spike-sorting sequence | false alarms | Hurt |
| Reduce kernel shift-coherence | Bilinear Lasso SaSD · Synthetic SaSD | successful recovery | Helped |
Closest papers by meaning
- Geometry and Symmetry in Short-and-Sparse Deconvolution (ICML 2019 · similarity 0.88)
- Structured Local Minima in Sparse Blind Deconvolution (NeurIPS 2018 · similarity 0.88)
- On the Global Geometry of Sphere-Constrained Sparse Blind Deconvolution (CVPR 2017 · similarity 0.86)
- A Nonconvex Approach for Exact and Efficient Multichannel Sparse Blind Deconvolution (NeurIPS 2019 · similarity 0.86)
- LeMoNADe: Learned Motif and Neuronal Assembly Detection in calcium imaging videos (ICLR 2019 · similarity 0.85)
- Sparse blind deconvolution via thresholded Wirtinger flow (NeurIPS 2026 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.