Machine learning and deep learning have shown great promise in mobile sensing applications, including Human Activity Recognition. However, the performance of such models in real-world settings largely depends on the availability of large datasets that captures diverse behaviors. Recently, studies in computer vision and natural language processing have shown that leveraging massive amounts of unlabeled data enables performance on par with state-of-the-art supervised models. In this work, we present SelfHAR, a semi-supervised model that effectively learns to leverage unlabeled mobile sensing datasets to complement small labeled datasets. Our approach combines teacher-student self-training, which distills the knowledge of unlabeled and labeled datasets while allowing for data augmentation, and multi-task self-supervision, which learns robust signal-level representations by predicting distorted versions of the input. We evaluated SelfHAR on various HAR datasets and showed state-of-the-art performance over supervised and previous semi-supervised approaches, with up to 12% increase in F1 score using the same number of model parameters at inference. Furthermore, SelfHAR is data-efficient, reaching similar performance using up to 10 times less labeled data compared to supervised approaches. Our work not only achieves state-of-the-art performance in a diverse set of HAR datasets, but also sheds light on how pre-training tasks may affect downstream performance.
[1]
J. Perry,et al.
Association of Genetic Variants Related to Gluteofemoral vs Abdominal Fat Distribution With Type 2 Diabetes, Coronary Disease, and Cardiovascular Risk Factors
,
2018,
JAMA.
[2]
N. Wareham,et al.
Estimation of Physical Activity Energy Expenditure during Free-Living from Wrist Accelerometry in UK Adults
,
2016,
PloS one.
[3]
Søren Brage,et al.
Impact of study design on development and evaluation of an activity-type classifier.
,
2013,
Journal of applied physiology.
[4]
Davide Anguita,et al.
A Public Domain Dataset for Human Activity Recognition using Smartphones
,
2013,
ESANN.
[5]
Geoffrey E. Hinton,et al.
Visualizing Data using t-SNE
,
2008
.