Quantum circuit improves scarce physiological signal data augmentation accuracy
Conditional Quantum Flow Matching for Data-Scarce Physiological Signal Augmentation
Machine LearningMultimedia
Summary
Physiological signals like brain waves are hard to classify well when there isn’t much labeled data. The authors propose a new quantum computing method that begins with more meaningful information tailored to the class to generate better synthetic data. This method uses fewer parameters and improves accuracy on a brain-computer interface task compared to previous quantum models that start from noise. Where simpler prior models fail, this method recovers performance better by learning from data transferred from other people.
What this means in practice
- •For healthcare data engineers: Enhance training data for physiological signal analysis when labeled samples are limited, improving model accuracy for brain-computer interfaces.
- •For wearable device developers: Create better synthetic physiological data for device calibration across users, reducing dependence on large labeled datasets.
Authors
Chi-Sheng Chen, Samuel Yen-Chi Chen
Abstract
Generative augmentation is a standard remedy for label scarcity in physiological signal classification, but existing quantum generative models start from uninformative noise, ignoring class structure that is already available. We propose Conditional Quantum Flow Matching (CQFM): a single 306-parameter circuit, conditioned on both flow time and class label, transports a compact class-conditional prior toward the target distribution. Quantum flow matching as published is unconditional, so this is to our knowledge the first conditional one, and the first EEG augmentation on a parameterized quantum circuit. A nonnegative spectral embedding removes the need for tomography at readout. On BCI Competition IV-2a, starting from a prior rather than noise is worth $+5.1$ accuracy points over QuDDPM (9/9 subjects), though at that operating point a class-conditional Gaussian matches CQFM. Where the prior fails the transport earns its keep: given one transferred from other subjects it regains $+7.2$ TSTR points (9/9).