Power in Unity: Combining in-Domain and out-of-Domain Pre-Training Strategies for EEG-Based Person Identification

Christos Garoufis; Marios Glytsos; Ioanna Chourdaki; Panagiotis Paraskevas Filntisis; Petros Maragos

ICASSP 2025accepted0 citations

Power in Unity: Combining in-Domain and out-of-Domain Pre-Training Strategies for EEG-Based Person Identification

Christos Garoufis, Marios Glytsos, Ioanna Chourdaki, Panagiotis Paraskevas Filntisis, Petros Maragos

Abstract

We present the NTUA-IRAL team’s solution for the Person Identification track of the Signal Processing EEG-Music Emotion Recognition Grand Challenge, hosted at ICASSP. Our approach employs an ensemble of three CNNs, each pretrained using a distinct strategy: contrastive pre-training, traditional ImageNet pre-training, and task-specific pre-training on a publicly available EEG dataset. This diverse pre-training regimen enabled our models to achieve a test set accuracy of 100%, earning third place in the challenge subtrack.

BibTeX

@inproceedings{icassp2025_powerinunitycomb,
  title = {Power in Unity: Combining in-Domain and out-of-Domain Pre-Training Strategies for EEG-Based Person Identification},
  author = {Christos Garoufis and Marios Glytsos and Ioanna Chourdaki and Panagiotis Paraskevas Filntisis and Petros Maragos},
  booktitle = {ICASSP 2025},
  year = {2025}
}