PiRL: Participant-Invariant Representation Learning for Healthcare
This work addresses the challenge of improving generic models for healthcare applications where person-specific models are impractical, offering an incremental improvement in accuracy for detecting conditions like sleep apnea and stress.
The paper tackled the problem of individual heterogeneity causing performance gaps between generic and person-specific models in healthcare by proposing PiRL, a representation learning framework that uses MMD loss, domain-adversarial training, and triplet loss to learn participant-invariant representations, resulting in around a 5% increase in accuracy on sleep apnea and stress detection tasks.
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.