PiRL: Participant-Invariant Representation Learning for Healthcare Using Maximum Mean Discrepancy and Triplet Loss

Abstract

Due to individual heterogeneity, person-specific models are usually achieving better performance than generic (one-size-fits-all) models in data-driven health applications. However, generic models are usually preferable in real-world applications, due to the difficulties of developing person-specific models, such as new-user-adaptation issues and system complexities. To improve the performance of generic models, we propose a Participant-invariant Representation Learning (PiRL) framework, which utilizes maximum mean discrepancy (MMD) loss and domain-adversarial training to encourage the model to learn participant-invariant representations. Further, to avoid trivial solutions in the learned representations, a triplet loss based constraint is used for the model to learn the label-distinguishable embeddings. The proposed framework is evaluated on two public datasets (CLAS and Apnea-ECG), and significant performance improvements are achieved compared to the baseline models.

TarDis: Achieving Robust and Structured Disentanglement of Multiple Covariates

“…Existing invariant representation learning methods often fail for continuous domain problems, an area that is significantly underexplored yet critically important [7, 96, 98]. Examples include patient monitoring systems where physiological spurious data varies daily and across activities [17], finance, where models predicting stock prices or market trends must generalize across varying economic conditions and times [34], and climate modeling, where models use invariant learning to forecast weather or long-term climate changes across diverse locations and time periods [10]. Existing methods are generally designed for discrete categorical domains and struggle with the continuous nature of many real-world tasks.…”