PiRL: Participant-Invariant Representation Learning for Healthcare

TL;DR

This paper introduces PiRL, a framework that learns participant-invariant representations to improve the accuracy of generic health models across individuals, using MMD, adversarial training, and triplet loss.

Contribution

The paper presents a novel participant-invariant representation learning framework combining MMD, adversarial training, and triplet loss for health data analysis.

Findings

Achieved around 5% accuracy improvement over baseline models.

Validated on datasets for sleep apnea and stress detection.

Demonstrated effectiveness of participant-invariant representations in health applications.

Abstract

Due to individual heterogeneity, performance gaps are observed between generic (one-size-fits-all) models and person-specific models in data-driven health applications. However, in real-world applications, generic models are usually more favorable due to new-user-adaptation issues and system complexities, etc. To improve the performance of the generic model, we propose a representation learning framework that learns participant-invariant representations, named PiRL. The proposed framework utilizes maximum mean discrepancy (MMD) loss and domain-adversarial training to encourage the model to learn participant-invariant representations. Further, a triplet loss, which constrains the model for inter-class alignment of the representations, is utilized to optimize the learned representations for downstream health applications. We evaluated our frameworks on two public datasets related to physical and mental health, for detecting sleep apnea and stress, respectively. As preliminary results, we found the proposed approach shows around a 5% increase in accuracy compared to the baseline.