DIVER-0 : A Fully Channel Equivariant EEG Foundation Model

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Han, Danny Dongyeop, Lee, Ahhyun Lucy, Lee, Taeyang, Gwon, Yonghyeon, Lee, Sebin, Lee, Seongjin, Park, David Keetae, Yoo, Shinjae, Cha, Jiook, Chung, Chun Kee
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866908455786250240
author Han, Danny Dongyeop
Lee, Ahhyun Lucy
Lee, Taeyang
Gwon, Yonghyeon
Lee, Sebin
Lee, Seongjin
Park, David Keetae
Yoo, Shinjae
Cha, Jiook
Chung, Chun Kee
author_facet Han, Danny Dongyeop
Lee, Ahhyun Lucy
Lee, Taeyang
Gwon, Yonghyeon
Lee, Sebin
Lee, Seongjin
Park, David Keetae
Yoo, Shinjae
Cha, Jiook
Chung, Chun Kee
contents Electroencephalography (EEG) is a non-invasive technique widely used in brain-computer interfaces and clinical applications, yet existing EEG foundation models face limitations in modeling spatio-temporal brain dynamics and lack channel permutation equivariance, preventing robust generalization across diverse electrode configurations. To address these challenges, we propose DIVER-0, a novel EEG foundation model that demonstrates how full spatio-temporal attention-rather than segregated spatial or temporal processing-achieves superior performance when properly designed with Rotary Position Embedding (RoPE) for temporal relationships and binary attention biases for channel differentiation. We also introduce Sliding Temporal Conditional Positional Encoding (STCPE), which improves upon existing conditional positional encoding approaches by maintaining both temporal translation equivariance and channel permutation equivariance, enabling robust adaptation to arbitrary electrode configurations unseen during pretraining. Experimental results demonstrate that DIVER-0 achieves competitive performance with only 10% of pretraining data while maintaining consistent results across all channel permutation conditions, validating its effectiveness for cross-dataset generalization and establishing key design principles for handling the inherent heterogeneity of neural recording setups.
format Preprint
id arxiv_https___arxiv_org_abs_2507_14141
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DIVER-0 : A Fully Channel Equivariant EEG Foundation Model
Han, Danny Dongyeop
Lee, Ahhyun Lucy
Lee, Taeyang
Gwon, Yonghyeon
Lee, Sebin
Lee, Seongjin
Park, David Keetae
Yoo, Shinjae
Cha, Jiook
Chung, Chun Kee
Signal Processing
Artificial Intelligence
Machine Learning
Electroencephalography (EEG) is a non-invasive technique widely used in brain-computer interfaces and clinical applications, yet existing EEG foundation models face limitations in modeling spatio-temporal brain dynamics and lack channel permutation equivariance, preventing robust generalization across diverse electrode configurations. To address these challenges, we propose DIVER-0, a novel EEG foundation model that demonstrates how full spatio-temporal attention-rather than segregated spatial or temporal processing-achieves superior performance when properly designed with Rotary Position Embedding (RoPE) for temporal relationships and binary attention biases for channel differentiation. We also introduce Sliding Temporal Conditional Positional Encoding (STCPE), which improves upon existing conditional positional encoding approaches by maintaining both temporal translation equivariance and channel permutation equivariance, enabling robust adaptation to arbitrary electrode configurations unseen during pretraining. Experimental results demonstrate that DIVER-0 achieves competitive performance with only 10% of pretraining data while maintaining consistent results across all channel permutation conditions, validating its effectiveness for cross-dataset generalization and establishing key design principles for handling the inherent heterogeneity of neural recording setups.
title DIVER-0 : A Fully Channel Equivariant EEG Foundation Model
topic Signal Processing
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2507.14141