mlsanityData to 7 Oct 2026

A Sliced-Wasserstein Framework on Correlation Matrices for EEG Decoding

arXiv 2606.06104 ↗First posted 4 Jun 2026cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Electroencephalography (EEG) offers noninvasive, millisecond resolution recordings of neuronal activity and is widely used in neuroscience and healthcare. Many EEG decoding pipelines rely on covariance descriptors for their robustness to noise, but such representations are sensitive to channel-wise scaling. Recent studies have therefore advocated full-rank correlation matrices as a scale-invariant alternative for EEG decoding. In this paper, we study Sliced-Wasserstein (SW) discrepancies for probability distributions on the manifold of full-rank correlation matrices. We adopt the pullback-Euclidean formulation of SW, referred to as Pullback Euclidean Metric Sliced-Wasserstein (PEMSW), and instantiate it under two recently introduced correlation geometries, textiti.e., the Off-Log Metric (OLM) and Log-Scaled Metric (LSM). This yields two Correlation Sliced-Wasserstein (CorSW) discrepancies with closed-form slicing coordinates and efficient computation through one-dimensional Wasserstein distances. Building on CorSW, we further develop a domain generalization (DG) framework for EEG decoding. Experiments on three EEG datasets demonstrate improved generalization under distribution shifts, with low training overhead and no additional inference cost. The source code is available at github.com/ChenHu-ML/CorSW.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 33 of its 80 references that are in the map. First posted on arXiv 4 Jun 2026.

Builds on

33 of 80 references.

References · 33

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2013Sinkhorn Distances: Lightspeed Computation of Optimal Transportation Distancescited by 2,277
  3. 2016EEGNet: A Compact Convolutional Network for EEG-based Brain-Computer Interfacescited by 616
  4. 2021RobustNet: Improving Domain Generalization in Urban-Scene Segmentation via Instance Selective Whiteningcited by 213
  5. 2016A Riemannian Network for SPD Matrix Learningcited by 162
  6. 2019Subspace Robust Wasserstein Distancescited by 89
  7. 2019Learning with minibatch Wasserstein : asymptotic and gradient propertiescited by 83
  8. 2020Statistical and Topological Properties of Sliced Probability Divergencescited by 83
  9. 2019Riemannian batch normalization for SPD neural networkscited by 72
  10. 2020EEG-TCNet: An Accurate Temporal Convolutional Network for Embedded Motor-Imagery Brain-Machine Interfacescited by 62
  11. 2016Building Deep Networks on Grassmann Manifoldscited by 58
  12. 2021FBCNet: A Multi-view Convolutional Neural Network for Brain-Computer Interfacecited by 54
21 more
  1. 2022SPD domain-specific batch normalization to crack interpretable unsupervised domain adaptation in EEGcited by 53
  2. 2019Asymptotic Guarantees for Learning Generative Models with the Sliced-Wasserstein Distancecited by 43
  3. 2022MAtt: A Manifold Attention Network for EEG Decodingcited by 43
  4. 2021Automatic Sleep Staging of EEG Signals: Recent Development, Challenges, and Future Directionscited by 41
  5. 2024Sliced-Wasserstein Distances and Flows on Cartan-Hadamard Manifoldscited by 28
  6. 2018ChronoNet: A Deep Recurrent Neural Network for Abnormal EEG Identificationcited by 27
  7. 2023Decompose, Adjust, Compose: Effective Normalization by Playing with Frequency for Domain Generalizationcited by 25
  8. 2022Riemannian Local Mechanism for SPD Neural Networkscited by 22
  9. 2023Sliced-Wasserstein on Symmetric Positive Definite Matrices for M/EEG Signalscited by 19
  10. 2022Hyperbolic Sliced-Wasserstein via Geodesic and Horospherical Projectionscited by 16
  11. 2022Graph Neural Networks on SPD Manifolds for Motor Imagery Classification: A Perspective from the Time-Frequency Analysiscited by 15
  12. 2023Adaptive Log-Euclidean Metrics for SPD Matrix Learningcited by 13
  13. 2024SPDIM: Source-Free Unsupervised Conditional and Label Shift Adaptation in EEGcited by 12
  14. 2016A Tutorial on Regularized Partial Correlation Networkscited by 11
  15. 2023Riemannian Multinomial Logistics Regression for SPD Neural Networkscited by 11
  16. 2024Understanding Matrix Function Normalizations in Covariance Pooling through the Lens of Riemannian Geometrycited by 9
  17. 2024A Lie Group Approach to Riemannian Batch Normalizationcited by 8
  18. 2020A New Parametrization of Correlation Matricescited by 7
  19. 2021Statistical and Topological Properties of Gaussian Smoothed Sliced Probability Divergencescited by 7
  20. 2025Learning to Normalize on the SPD Manifold under Bures-Wasserstein Geometrycited by 7
  21. 2022Deep Optimal Transport for Domain Adaptation on SPD Manifoldscited by 6

Built on by

Not cited yet.