MPNet: A Robust and Efficient Manifold Pooling Network for Multi-Rhythm EEG Signal Decoding

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Cai, Guoqing, Zeng, Kai, Huang, Shoulin, Ma, Ting
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866917466475593728
author Cai, Guoqing
Zeng, Kai
Huang, Shoulin
Ma, Ting
author_facet Cai, Guoqing
Zeng, Kai
Huang, Shoulin
Ma, Ting
contents Deep Riemannian networks provide a powerful framework for Electroencephalography (EEG) decoding, but their practical applications are severely constrained. Accurately decoding EEG signals requires modeling complex temporal dynamics across multiple rhythms, which results in high-dimensional Riemannian inputs and significant computational costs. To address this, we propose the Manifold Pooling Network (MPNet). MPNet uses a rhythm-adaptive convolutional frontend to extract comprehensive time-frequency representations and generate multi-view Riemannian nodes. A novel manifold node pooling layer is then proposed to aggregate these nodes into a single fusion node with a fixed size, enabling the following deep Riemannian network to process it with greatly reduced costs. Experiments on two public EEG datasets show that MPNet achieves state-of-the-art accuracy, runs up to 10 times faster than the comparable Riemannian model, and maintains robust performance under limited-data conditions. These findings highlight MPNet's practicality and efficiency for real-world EEG applications.
format Preprint
id arxiv_https___arxiv_org_abs_2605_05212
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle MPNet: A Robust and Efficient Manifold Pooling Network for Multi-Rhythm EEG Signal Decoding
Cai, Guoqing
Zeng, Kai
Huang, Shoulin
Ma, Ting
Signal Processing
Human-Computer Interaction
Machine Learning
Deep Riemannian networks provide a powerful framework for Electroencephalography (EEG) decoding, but their practical applications are severely constrained. Accurately decoding EEG signals requires modeling complex temporal dynamics across multiple rhythms, which results in high-dimensional Riemannian inputs and significant computational costs. To address this, we propose the Manifold Pooling Network (MPNet). MPNet uses a rhythm-adaptive convolutional frontend to extract comprehensive time-frequency representations and generate multi-view Riemannian nodes. A novel manifold node pooling layer is then proposed to aggregate these nodes into a single fusion node with a fixed size, enabling the following deep Riemannian network to process it with greatly reduced costs. Experiments on two public EEG datasets show that MPNet achieves state-of-the-art accuracy, runs up to 10 times faster than the comparable Riemannian model, and maintains robust performance under limited-data conditions. These findings highlight MPNet's practicality and efficiency for real-world EEG applications.
title MPNet: A Robust and Efficient Manifold Pooling Network for Multi-Rhythm EEG Signal Decoding
topic Signal Processing
Human-Computer Interaction
Machine Learning
url https://arxiv.org/abs/2605.05212