XAI-BayesHAR: A novel Framework for Human Activity Recognition with Integrated Uncertainty and Shapely Values
This work addresses practical deployment challenges in smart homes and healthcare by providing uncertainty analysis and model compression for activity recognition systems, though it appears incremental as it builds on existing Bayesian and XAI methods.
The paper tackles the problem of improving accuracy and uncertainty estimation in IMU-based human activity recognition by proposing XAI-BayesHAR, a Bayesian framework that integrates Kalman filtering for tracking feature embeddings and uncertainty, resulting in enhanced classification accuracy and out-of-distribution detection.
Human activity recognition (HAR) using IMU sensors, namely accelerometer and gyroscope, has several applications in smart homes, healthcare and human-machine interface systems. In practice, the IMU-based HAR system is expected to encounter variations in measurement due to sensor degradation, alien environment or sensor noise and will be subjected to unknown activities. In view of practical deployment of the solution, analysis of statistical confidence over the activity class score are important metrics. In this paper, we therefore propose XAI-BayesHAR, an integrated Bayesian framework, that improves the overall activity classification accuracy of IMU-based HAR solutions by recursively tracking the feature embedding vector and its associated uncertainty via Kalman filter. Additionally, XAI-BayesHAR acts as an out of data distribution (OOD) detector using the predictive uncertainty which help to evaluate and detect alien input data distribution. Furthermore, Shapley value-based performance of the proposed framework is also evaluated to understand the importance of the feature embedding vector and accordingly used for model compression