ADVANCING HUMAN ACTIVITY RECOGNITION THROUGH WEAKLY AND SELF-SUPERVISED LEARNING: TOWARDS LABEL-EFFICIENT WEARABLE INTELLIGENCE

With the use of wearable devices, Human Activity Recognition (HAR) has gained more value owing to itsrelevance in healthcare, fitness, and human-computer interaction. The conventional methods of supervisedlearning of HAR make use of large labeled datasets, which are costly and time-intensive to acquire. Newdevelopments in weakly and self supervised learning provide compelling options by minimizing the use of largescale labeling whilst preserving strong recognition. Methods that are weakly supervised utilize limited or coarsegrained labels to learn multi-task representations to learn complex patterns of activities without the exhaustiveannotation. Self-supervised methods build on the natural form of sensor data to train useful embeddings of timeseries that are unlabeled, which can be used to complete downstream HAR tasks.Continuity-based learning with embedding strategies have demonstrated a great enhancement in labeleffectiveness, enabling scalable and viable solutions of HAR. Even with these developments, there are still severalconcerns such as variations in sensor locations, inter-subject variations, and temporal variations, and it isimperative to have strong and flexible representation learning models. Further, there are new solutions thatconsider other sensing modalities, including WiFi signals and biomedical sensors that exhibit how well selfsupervised learning can be used in multi-user and clinical settings.The paper will review and synthesize recent advances in weakly and self-supervised learning in HAR, particularly,label-efficient wearable intelligence. Through the analysis of the state-of-the-art techniques and their applications,it gives insights on future tendencies of the scalable, low-label, and adaptive HAR systems. Wearable intelligencecan be extended through integration of weak and self-supervised learning which would allow individual but realtime monitoring of activities with minimum manual annotation.