Comparing Self-Supervised Learning Techniques for Wearable Human Activity Recognition

TL;DR

This paper compares different self-supervised learning techniques for wearable human activity recognition, demonstrating that Masked Auto Encoder significantly outperforms other methods in accuracy with limited labeled data.

Contribution

It adapts and evaluates three SSL techniques for HAR and shows that MAE outperforms contrastive and generative methods, providing a new state-of-the-art approach.

Findings

MAE outperforms other SSL methods in HAR accuracy

Transformer architectures improve recognition rates

Code and models are publicly available for research

Abstract

Human Activity Recognition (HAR) based on the sensors of mobile/wearable devices aims to detect the physical activities performed by humans in their daily lives. Although supervised learning methods are the most effective in this task, their effectiveness is constrained to using a large amount of labeled data during training. While collecting raw unlabeled data can be relatively easy, annotating data is challenging due to costs, intrusiveness, and time constraints. To address these challenges, this paper explores alternative approaches for accurate HAR using a limited amount of labeled data. In particular, we have adapted recent Self-Supervised Learning (SSL) algorithms to the HAR domain and compared their effectiveness. We investigate three state-of-the-art SSL techniques of different families: contrastive, generative, and predictive. Additionally, we evaluate the impact of the underlying neural network on the recognition rate by comparing state-of-the-art CNN and transformer architectures. Our results show that a Masked Auto Encoder (MAE) approach significantly outperforms other SSL approaches, including SimCLR, commonly considered one of the best-performing SSL methods in the HAR domain. The code and the pre-trained SSL models are publicly available for further research and development.