Research map / Time series / Causal and generative time series
Learning the Morphology of Brain Signals Using Alpha-Stable Convolutional Sparse Coding
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 heavy-tailed α-stable likelihood instead of Gaussian likelihood | αCSC · Synthetic signals with corrupted trials | atom recovery robustness | Helped |
| Use Gaussian CSC (α=2) instead of heavy-tailed αCSC (ablation) | CSC · Synthetic signals with corrupted trials | atom recovery robustness | Hurt |
| Use MoTIF with activations of varying amplitudes | MoTIF · Artifact-free synthetic signals | atom recovery | Hurt |
| Use the proposed quasi-Newton M-step instead of ADMM-based CSC solvers | CSC M-step · Synthetic M-step performance benchmark | convergence speed | Helped |
| Use L-BFGS-B for the activation update instead of ISTA or FISTA (ablation) | CSC M-step · Synthetic activation-subproblem benchmark | convergence speed | Helped |
| Manually remove artifact-corrupted trials before fitting standard CSC (ablation) | CSC · Rodent striatal LFP | recovery of canonical atoms | Helped |
| Fit standard CSC to the full data without removing artifacts (ablation) | CSC · Rodent striatal LFP | recovery of canonical atoms | Hurt |
| Use αCSC with α=1.2 on full data rather than standard CSC | CSC · Rodent striatal LFP | recovery of cross-frequency coupling | Helped |
| Use α=2, making the likelihood Gaussian (ablation) | αCSC · Synthetic M-step benchmark | M-step convergence speed | No clear effect |
| Use α=2 standard CSC with the proposed optimizer on artifact-free epileptiform LFP (ablation) | αCSC optimization scheme · Rat cortex LFP | recovery of spike-like atoms | Helped |
Closest papers by meaning
- Multivariate Convolutional Sparse Coding for Electromagnetic Brain Signals (NeurIPS 2018 · similarity 0.89)
- Select-and-Sample for Spike-and-Slab Sparse Coding (NeurIPS 2016 · similarity 0.85)
- Rethinking the CSC Model for Natural Images (NeurIPS 2019 · similarity 0.85)
- Dynamical modeling of nonlinear latent factors in multiscale neural activity with real-time inference (NeurIPS 2025 · similarity 0.84)
- A sampling theory perspective on activations for implicit neural representations (ICML 2024 · similarity 0.83)
- FreSh: Frequency Shifting for Accelerated Neural Representation Learning (ICLR 2025 · similarity 0.83)
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.