Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

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

Debaditya Shome, Nasim Montazeri Ghahjaverestan, Ali Etemad

TL;DR

Through rigorous empirical evaluation, it is demonstrated 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.

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.

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

TL;DR

Through rigorous empirical evaluation, it is demonstrated 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.

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.
Paper Structure (21 sections, 6 equations, 5 figures, 4 tables)

This paper contains 21 sections, 6 equations, 5 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Wearable affect recognition
  4. Sleep and affect
  5. Prompt tuning
  6. Method
  7. Problem and solution overview
  8. Proposed approach
  9. Wearable time-series encoder
  10. Pre-training
  11. Tuning
  12. Experiment Setup
  13. Datasets
  14. Pre-training datasets
  15. Downstream Dataset
  16. ...and 6 more sections

Figures (5)

  • Figure 1: Our proposed NapTune framework.
  • Figure 2: The architecture of the wearable time-series encoder.
  • Figure 3: Impact of sleep on classification of different classes of mood with EDA (left), PPG (middle), and ECG (right).
  • Figure 4: Ablation of individual sleep-related measures.
  • Figure 5: Impact of the amount of data used for training with ECG+Sleep.