Research map / Time series / Long-term forecasting
Multivariate Convolutional Sparse Coding for Electromagnetic Brain Signals
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 multivariate rank-1 CSC instead of univariate CSC | Synthetic pattern recovery · Synthetic data | temporal-pattern recovery loss | Helped |
| increase the number of channels from one to five (ablation) | Multivariate rank-1 CSC · Synthetic data | temporal-pattern recovery | Helped |
| use rank-1 multivariate atoms rather than univariate atoms | Synthetic data with overlapping patterns · Synthetic data | pattern recovery | Helped |
| use LGCD instead of competing CSC solvers | Univariate CSC on somatosensory MEG · Somatosensory MEG, long signals | runtime | Mixed |
| use LGCD instead of multivariate ADMM | Multivariate CSC · Somatosensory MEG, P=5 | runtime | Helped |
| use LGCD with small regularization parameters (ablation) | CSC solvers · Somatosensory MEG, shorter signals | runtime | Hurt |
| initialize atoms from random signal chunks projected to rank 1 with SVD instead of Gaussian white noise (ablation) | Multivariate CSC rank-1 · Somatosensory MEG | initial lambda_max: 1/20 | Helped |
| add the rank-1 spatial-temporal atom constraint | Multivariate CSC · EEG/MEG signals | computational complexity | Helped |
| use precomputations in the optimization (ablation) | Multivariate CSC · Somatosensory MEG with varying channel count | runtime scaling with channels | Helped |
| fit multivariate rank-1 atoms to MEG recordings | Multivariate CSC rank-1 · MNE somatosensory MEG | goodness of fit: 59.3% | Helped |
Closest papers by meaning
- BrainOmni: A Brain Foundation Model for Unified EEG and MEG Signals (NeurIPS 2025 · similarity 0.85)
- Brain decoding: toward real-time reconstruction of visual perception (ICLR 2024 · similarity 0.84)
- CSBrain: A Cross-scale Spatiotemporal Brain Foundation Model for EEG Decoding (NeurIPS 2025 · similarity 0.84)
- Decoding Natural Images from EEG for Object Recognition (ICLR 2024 · similarity 0.84)
- Visual Decoding and Reconstruction via EEG Embeddings with Guided Diffusion (NeurIPS 2024 · similarity 0.83)
- Brain-OF: An Omnifunctional Foundation Model for fMRI, EEG and MEG (NeurIPS 2026 · 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.