Transformer-Based Contrastive Meta-Learning For Low-Resource Generalizable Activity Recognition
This addresses the challenge of costly data collection and labeling in human activity recognition for applications requiring adaptation to new users or environments, representing an incremental improvement with a hybrid method.
The paper tackles the problem of generalizing human activity recognition models across diverse users and scenarios under low-resource conditions, proposing TACO, a transformer-based contrastive meta-learning approach that achieves notably better performance in various low-resource distribution shift scenarios.
Deep learning has been widely adopted for human activity recognition (HAR) while generalizing a trained model across diverse users and scenarios remains challenging due to distribution shifts. The inherent low-resource challenge in HAR, i.e., collecting and labeling adequate human-involved data can be prohibitively costly, further raising the difficulty of tackling DS. We propose TACO, a novel transformer-based contrastive meta-learning approach for generalizable HAR. TACO addresses DS by synthesizing virtual target domains in training with explicit consideration of model generalizability. Additionally, we extract expressive feature with the attention mechanism of Transformer and incorporate the supervised contrastive loss function within our meta-optimization to enhance representation learning. Our evaluation demonstrates that TACO achieves notably better performance across various low-resource DS scenarios.