Universal Physiological Representation Learning with Soft-Disentangled Rateless Autoencoders
This addresses the problem of unstable physiological signal interpretation for HCI applications, offering an incremental improvement in transfer learning accuracy.
The paper tackles the challenge of physiological biosignal variability across users and sessions in human-computer interaction by proposing a method using adversarial feature encoding with Rateless Autoencoders to learn disentangled, nuisance-robust representations. It achieves up to an 11.6% improvement in average subject-transfer classification accuracy in cross-subject evaluations.
Human computer interaction (HCI) involves a multidisciplinary fusion of technologies, through which the control of external devices could be achieved by monitoring physiological status of users. However, physiological biosignals often vary across users and recording sessions due to unstable physical/mental conditions and task-irrelevant activities. To deal with this challenge, we propose a method of adversarial feature encoding with the concept of a Rateless Autoencoder (RAE), in order to exploit disentangled, nuisance-robust, and universal representations. We achieve a good trade-off between user-specific and task-relevant features by making use of the stochastic disentanglement of the latent representations by adopting additional adversarial networks. The proposed model is applicable to a wider range of unknown users and tasks as well as different classifiers. Results on cross-subject transfer evaluations show the advantages of the proposed framework, with up to an 11.6% improvement in the average subject-transfer classification accuracy.