Q-DIVER: Integrated Quantum Transfer Learning and Differentiable Quantum Architecture Search with EEG Data

TL;DR

Q-DIVER introduces a hybrid quantum-classical framework with differentiable architecture search, achieving competitive EEG classification performance with significantly fewer parameters.

Contribution

It presents a novel quantum transfer learning approach with differentiable quantum architecture search for EEG data classification.

Findings

Quantum classifier achieves 63.49% F1 score on PhysioNet dataset.

Uses approximately 50 times fewer task-specific parameters than classical models.

Validates quantum transfer learning as a parameter-efficient method.

Abstract

Integrating quantum circuits into deep learning pipelines remains challenging due to heuristic design limitations. We propose Q-DIVER, a hybrid framework combining a large-scale pretrained EEG encoder (DIVER-1) with a differentiable quantum classifier. Unlike fixed-ansatz approaches, we employ Differentiable Quantum Architecture Search to autonomously discover task-optimal circuit topologies during end-to-end fine-tuning. On the PhysioNet Motor Imagery dataset, our quantum classifier achieves predictive performance comparable to classical multi-layer perceptrons (Test F1: 63.49\%) while using approximately \textbf{50$\times$ fewer task-specific head parameters} (2.10M vs. 105.02M). These results validate quantum transfer learning as a parameter-efficient strategy for high-dimensional biological signal processing.