First steps to the classification of ERPs using quantum computation

Grégoire Cattan (1) , Anton Andreev (2)

1 IBM - Cloud and Cognitive Software
2 GIPSA-Services - GIPSA-Services

In this study, we investigate the performance of a quantum-enhanced support vector classifier for the binary classification of event-related potentials with electroencephalography. The resulting balanced accuracy on the training data was 83.17 %, hence comparing with the results of prior arts. This suggests that QSVC is a promising technology for the classification of event-related potentials.

Mots clés: Machine learning; Event-related Potentials ERP; Brain-computer interface BCI; Electroencephalograhy EEG; Quantum computing

Mots clés: Potentiels Evoqués Cognitifs PEC; Interface cerveau-machine ICM; Electroencephalographie EEG; Calcul quantique; Apprentissage supervisé

Domaines: Neurosciences; Physique Quantique [quant-ph]; Informatique; Psychologie; Interface homme-machine [cs.HC]; Intelligence artificielle [cs.AI]

Grégoire Cattan, Anton Andreev. First steps to the classification of ERPs using quantum computation. NTB Berlin 2022 - International Forum on Neural Engineering & Brain Technologies, May 2022, Berlin, Germany. ⟨hal-03672246⟩