NapTune: Efficient Model Tuning for Mood Classification using Previous Night's Sleep Measures along with Wearable Time-series

TL;DR

NapTune is a novel prompt-tuning framework that enhances wearable-based mood classification by integrating sleep measures from the previous night, leading to improved accuracy and sample efficiency across multiple physiological signals.

Contribution

The paper introduces NapTune, a new prompt-tuning approach that incorporates sleep data into pre-trained wearable time-series models for better mood recognition.

Findings

NapTune improves mood classification accuracy across ECG, PPG, and EDA signals.

Inclusion of sleep data enhances sample efficiency of the model.

NapTune outperforms baseline and unimodal methods in experiments.

Abstract

Sleep is known to be a key factor in emotional regulation and overall mental health. In this study, we explore the integration of sleep measures from the previous night into wearable-based mood recognition. To this end, we propose NapTune, a novel prompt-tuning framework that utilizes sleep-related measures as additional inputs to a frozen pre-trained wearable time-series encoder by adding and training lightweight prompt parameters to each Transformer layer. Through rigorous empirical evaluation, we demonstrate that the inclusion of sleep data using NapTune not only improves mood recognition performance across different wearable time-series namely ECG, PPG, and EDA, but also makes it more sample-efficient. Our method demonstrates significant improvements over the best baselines and unimodal variants. Furthermore, we analyze the impact of adding sleep-related measures on recognizing different moods as well as the influence of individual sleep-related measures.