Temporally Multi-Scale Sparse Self-Attention for Physical Activity Data Imputation
Abstract
Wearable sensors enable health researchers to continuously collect data pertaining to the physiological state of individuals in real-world settings. However, such data can be subject to extensive missingness due to a complex combination of factors. In this work, we study the problem of imputation of missing step count data, one of the most ubiquitous forms of wearable sensor data. We construct a novel and large scale data set consisting of a training set with over 3 million hourly step count observations and a test set with over 2.5 million hourly step count observations. We propose a domain knowledge-informed sparse self-attention model for this task that captures the temporal multi-scale nature of step-count data. We assess the performance of the model relative to baselines and conduct ablation studies to verify our specific model designs.
Cite
Text
Wei et al. "Temporally Multi-Scale Sparse Self-Attention for Physical Activity Data Imputation." Proceedings of the fifth Conference on Health, Inference, and Learning, 2024.Markdown
[Wei et al. "Temporally Multi-Scale Sparse Self-Attention for Physical Activity Data Imputation." Proceedings of the fifth Conference on Health, Inference, and Learning, 2024.](https://mlanthology.org/chil/2024/wei2024chil-temporally/)BibTeX
@inproceedings{wei2024chil-temporally,
title = {{Temporally Multi-Scale Sparse Self-Attention for Physical Activity Data Imputation}},
author = {Wei, Hui and Xu, Maxwell A and Samplawski, Colin and Rehg, James Matthew and Kumar, Santosh and Marlin, Benjamin},
booktitle = {Proceedings of the fifth Conference on Health, Inference, and Learning},
year = {2024},
pages = {137-154},
volume = {248},
url = {https://mlanthology.org/chil/2024/wei2024chil-temporally/}
}