Research map / Probabilistic and scientific modeling / Dynamical systems
Manifold-regression to predict from MEG/EEG brain signals without source modeling
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 Riemannian covariance vectorization rather than sensor-space log-diagonal features | MEG age prediction · Cam-CAN MEG | MAE: 8.1y vs. worse | Helped |
| Use MRI-informed MNE source projection instead of data-driven Riemannian projection | MEG age prediction · Cam-CAN MEG | MAE: 8.1y -> 7.4y | Helped |
| Add supervised spatial filtering to log-diagonal features (ablation) | Log-diag · Cam-CAN MEG | improvement over identity log-diag: 73% | Helped |
| Use identity projection with Wasserstein vectorization instead of identity log-diag | MEG age prediction · Cam-CAN MEG | improvement over baseline: 85% | Helped |
| Use unsupervised projection with geometric vectorization instead of identity log-diag | MEG age prediction · Cam-CAN MEG | improvement over baseline: 96% | Helped |
| Use MNE projection instead of identity log-diag | MEG age prediction · Cam-CAN MEG | improvement over baseline: 95% | Helped |
| Change the projection strategy while using Riemannian geometry (ablation) | Riemannian methods · Cam-CAN MEG | MAE | No clear effect |
| Use supervised rather than unsupervised projection with non-Riemannian log-diag features (ablation) | Log-diag · Cam-CAN MEG | MAE | Helped |
| Use geometric vectorization rather than log-diagonal features under varying sensor mixing | Synthetic generative model · Synthetic data with subject-dependent mixing | MAE | Helped |
| Use Wasserstein vectorization rather than supervised log-diagonal features at high mixing variability | Synthetic generative model · Synthetic data with subject-dependent mixing | MAE | Helped |
| Use Wasserstein vectorization when the mixing matrix is orthogonal and noise tends to zero | Synthetic generative model · Synthetic data with square-root source-power relationship | out-of-sample prediction | Helped |
Closest papers by meaning
- Geodesic Optimization for Predictive Shift Adaptation on EEG data (NeurIPS 2024 · similarity 0.89)
- Riemannian Flow Matching for Brain Connectivity Matrices via Pullback Geometry (NeurIPS 2025 · similarity 0.88)
- Implicit Gaussian process representation of vector fields over arbitrary latent manifolds (ICLR 2024 · similarity 0.87)
- Exploring the Enigma of Neural Dynamics Through A Scattering-Transform Mixer Landscape for Riemannian Manifold (ICML 2024 · similarity 0.86)
- SPDIM: Source-Free Unsupervised Conditional and Label Shift Adaptation in EEG (ICLR 2025 · similarity 0.85)
- Geometry-Aware Similarity Metrics for Neural Representations on Riemannian and Statistical Manifolds (NeurIPS 2026 · similarity 0.84)
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.