Robust Feature Representation Using Multi-Task Learning for Human Activity Recognition

• Published in: Sensors (Basel, Switzerland) Vol. 24; no. 2; p. 681
• Main Authors: Azadi, Behrooz, Haslgrübler, Michael, Anzengruber-Tanase, Bernhard, Sopidis, Georgios, Ferscha, Alois
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 01.01.2024; MDPI
• Subjects: alpine skiing; Analysis; Classification; Computational linguistics; Datasets; Deep learning; human activity recognition; Language processing; Machine learning; multi-task learning; Natural language interfaces; representation learning; Sensors; Skis and skiing; wearable; Germany; deep learning; multi-task learning; wearable; alpine skiing; representation learning; human activity recognition

Learning underlying patterns from sensory data is crucial in the Human Activity Recognition (HAR) task to avoid poor generalization when coping with unseen data. A key solution to such an issue is representation learning, which becomes essential when input signals contain activities with similar pat...