Transfer Learning in Brain-Computer Interfaces with Adversarial Variational Autoencoders

We introduce adversarial neural networks for representation learning as a novel approach to transfer learning in brain-computer interfaces (BCIs). The proposed approach aims to learn subject-invariant representations by simultaneously training a conditional variational autoencoder (cVAE) and an adve...

Full description

Saved in:
Bibliographic Details
Published inInternational IEEE/EMBS Conference on Neural Engineering (Online) pp. 207 - 210
Main Authors Ozdenizci, Ozan, Wang, Ye, Koike-Akino, Toshiaki, Erdogmus, Deniz
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.03.2019
Subjects
adversarial networks
Brain modeling
Brain-computer interfaces
convolutional neural networks
Decoding
EEG
Electroencephalography
Feature extraction
representation learning
Training
Training data
transfer learning
variational autoencoders
Online AccessGet full text

Cover

More Information
Summary:We introduce adversarial neural networks for representation learning as a novel approach to transfer learning in brain-computer interfaces (BCIs). The proposed approach aims to learn subject-invariant representations by simultaneously training a conditional variational autoencoder (cVAE) and an adversarial network. We use shallow convolutional architectures to realize the cVAE, and the learned encoder is transferred to extract subject-invariant features from unseen BCI users' data for decoding. We demonstrate a proof-of-concept of our approach based on analyses of electroencephalographic (EEG) data recorded during a motor imagery BCI experiment.
ISSN:1948-3554
DOI:10.1109/NER.2019.8716897