Comparing Self-Supervised Learning Techniques for Wearable Human Activity Recognition

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Ek, Sannara, Presotto, Riccardo, Civitarese, Gabriele, Portet, François, Lalanda, Philippe, Bettini, Claudio
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910420904706048
author Ek, Sannara
Presotto, Riccardo
Civitarese, Gabriele
Portet, François
Lalanda, Philippe
Bettini, Claudio
author_facet Ek, Sannara
Presotto, Riccardo
Civitarese, Gabriele
Portet, François
Lalanda, Philippe
Bettini, Claudio
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2404_15331
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Comparing Self-Supervised Learning Techniques for Wearable Human Activity Recognition
Ek, Sannara
Presotto, Riccardo
Civitarese, Gabriele
Portet, François
Lalanda, Philippe
Bettini, Claudio
Signal Processing
Human-Computer Interaction
I.2
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.
title Comparing Self-Supervised Learning Techniques for Wearable Human Activity Recognition
topic Signal Processing
Human-Computer Interaction
I.2
url https://arxiv.org/abs/2404.15331