Semantic-Discriminative Mixup for Generalizable Sensor-based Cross-domain Activity Recognition

TL;DR

This paper introduces SDMix, a novel data augmentation method for sensor-based human activity recognition that improves model generalization across different domains without needing target domain data.

Contribution

The paper proposes a semantic-aware Mixup technique combined with a large margin loss to enhance cross-domain generalization in HAR models, addressing domain shift without target data.

Findings

SDMix outperforms state-of-the-art methods by 6% on average.

It improves cross-person, cross-dataset, and cross-position HAR accuracy.

The approach effectively handles semantic inconsistency and noisy labels.

Abstract

It is expensive and time-consuming to collect sufficient labeled data to build human activity recognition (HAR) models. Training on existing data often makes the model biased towards the distribution of the training data, thus the model might perform terribly on test data with different distributions. Although existing efforts on transfer learning and domain adaptation try to solve the above problem, they still need access to unlabeled data on the target domain, which may not be possible in real scenarios. Few works pay attention to training a model that can generalize well to unseen target domains for HAR. In this paper, we propose a novel method called Semantic-Discriminative Mixup (SDMix) for generalizable cross-domain HAR. Firstly, we introduce semantic-aware Mixup that considers the activity semantic ranges to overcome the semantic inconsistency brought by domain differences. Secondly, we introduce the large margin loss to enhance the discrimination of Mixup to prevent misclassification brought by noisy virtual labels. Comprehensive generalization experiments on five public datasets demonstrate that our SDMix substantially outperforms the state-of-the-art approaches with 6% average accuracy improvement on cross-person, cross-dataset, and cross-position HAR.