On Attention Models for Human Activity Recognition
This addresses a domain-specific issue for researchers and practitioners in human activity recognition using body-worn sensors, but it is incremental as it builds on an existing deep learning model.
The paper tackled the problem of modeling time-series data in human activity recognition with fixed temporal contexts that may not suit activities of varying durations, by introducing attention models to weight relevant temporal context, resulting in a significant increase in performance on benchmark datasets.
Most approaches that model time-series data in human activity recognition based on body-worn sensing (HAR) use a fixed size temporal context to represent different activities. This might, however, not be apt for sets of activities with individ- ually varying durations. We introduce attention models into HAR research as a data driven approach for exploring relevant temporal context. Attention models learn a set of weights over input data, which we leverage to weight the temporal context being considered to model each sensor reading. We construct attention models for HAR by adding attention layers to a state- of-the-art deep learning HAR model (DeepConvLSTM) and evaluate our approach on benchmark datasets achieving sig- nificant increase in performance. Finally, we visualize the learned weights to better understand what constitutes relevant temporal context.