Disentangled Adversarial Autoencoder for Subject-Invariant Physiological Feature Extraction

TL;DR

This paper introduces a disentangled adversarial autoencoder that enhances physiological feature extraction by learning universal representations, improving cross-user transferability and classification accuracy in biosignal processing.

Contribution

It proposes a novel adversarial autoencoder framework with disentangled representations and additional adversary networks for improved transfer learning across subjects and tasks.

Findings

Up to 8.8% improvement in classification accuracy.

Enhanced adaptability to diverse subjects.

Effective disentanglement of task-relevant and user-discriminative features.

Abstract

Recent developments in biosignal processing have enabled users to exploit their physiological status for manipulating devices in a reliable and safe manner. One major challenge of physiological sensing lies in the variability of biosignals across different users and tasks. To address this issue, we propose an adversarial feature extractor for transfer learning to exploit disentangled universal representations. We consider the trade-off between task-relevant features and user-discriminative information by introducing additional adversary and nuisance networks in order to manipulate the latent representations such that the learned feature extractor is applicable to unknown users and various tasks. Results on cross-subject transfer evaluations exhibit the benefits of the proposed framework, with up to 8.8% improvement in average accuracy of classification, and demonstrate adaptability to a broader range of subjects.