Embedded Inter-Subject Variability in Adversarial Learning for Inertial Sensor-Based Human Activity Recognition
This work addresses the generalization problem for HAR models to new individuals, which is a common issue in real-world applications of wearable sensors.
This paper tackles the challenge of inter-subject variability in Human Activity Recognition (HAR) using wearable inertial sensors. The authors propose a novel deep adversarial framework that integrates inter-subject variability into the adversarial task, leading to improved classification performance on three established HAR datasets using leave-one-subject-out cross-validation.
This paper addresses the problem of Human Activity Recognition (HAR) using data from wearable inertial sensors. An important challenge in HAR is the model's generalization capabilities to new unseen individuals due to inter-subject variability, i.e., the same activity is performed differently by different individuals. To address this problem, we propose a novel deep adversarial framework that integrates the concept of inter-subject variability in the adversarial task, thereby encouraging subject-invariant feature representations and enhancing the classification performance in the HAR problem. Our approach outperforms previous methods in three well-established HAR datasets using a leave-one-subject-out (LOSO) cross-validation. Further results indicate that our proposed adversarial task effectively reduces inter-subject variability among different users in the feature space, and it outperforms adversarial tasks from previous works when integrated into our framework. Code: https://github.com/FranciscoCalatrava/EmbeddedSubjectVariability.git