Improving Task-Incremental Human Activity Recognition with Plasticity Techniques

• F. G. Brabes da S. UNICAMP / H.IAAC
• W. G. S. Lima UNICAMP / H.IAAC
• A. Anjos da S. UNICAMP / H.IAAC
• E. L. Colombini UNICAMP / H.IAAC
• C. G. Ghedini UNICAMP / H.IAAC

DOI: https://doi.org/10.5753/kdmile.2024.244696

Resumo

Sensor-based Human Activity Recognition (HAR) has been applied across various domains, including healthcare monitoring, fitness tracking, and smart home systems. These applications require the ability to accurately detect and respond to a wide range of human activities, each with varying distributions, which imposes a significant challenge. The task-incremental learning paradigm can address this problem by enabling HAR systems to adapt to changes in distribution and learn new activities over time without forgetting the previously learned ones. Continual adaptation is essential for maintaining high performance, as it allows the system to effectively respond to these changes. Although several strategies in the continual learning literature have been evaluated for task-incremental scenarios in HAR, there is still room for improvement, as the results are not as good as those achieved with conventional approaches. This work proposes two new neuroplasticity-inspired techniques that can be integrated with any learning strategy. Inspired by the brain’s ability to reorganize and strengthen connections over time, these methods focus on enhancing the model’s flexibility and long-term knowledge retention. The proposed techniques were evaluated alongside the WA-ADB and WA-MDF strategies on well-known HAR datasets. Experimental results demonstrated that the new techniques significantly enhanced the models’ ability to retain knowledge, which holds significant potential for improving the robustness and longevity of HAR systems in real-world applications.

Palavras-chave: continual learning, human activity recognition (HAR), neuroplasticity, task-incremental learning

Referências

Anguita, D., Ghio, A., Oneto, L., Parra, X., and Reyes-Ortiz, J. L. A public domain dataset for human activity recognition using smartphones. In The European Symposium on Artificial Neural Networks, 2013.

Barshan, B. and Yüksek, M. C. Recognizing Daily and Sports Activities in Two Open Source Machine Learning Environments Using Body-Worn Sensor Units. The Computer Journal 57 (11): 1649–1667, 07, 2013.

Carta, A., Pellegrini, L., Cossu, A., Hemati, H., and Lomonaco, V. Avalanche: A pytorch library for deep continual learning. Journal of Machine Learning Research 24 (363): 1–6, 2023.

Jha, S., Schiemer, M., Zambonelli, F., and Ye, J. Continual learning in sensor-based human activity recognition: An empirical benchmark analysis. Information Sciences vol. 575, pp. 1–21, 2021.

Kim, B. and Kim, J. Adjusting decision boundary for class imbalanced learning. IEEE Access vol. PP, pp. 1–1, 04, 2020.

Lee, H., Grosse, R., Ranganath, R., and Ng, A. Unsupervised learning of hierarchical representations with convolutional deep belief networks. Commun. ACM vol. 54, pp. 95–103, 10, 2011.

Mann, H. B. and Whitney, D. R. On a test of whether one of two random variables is stochastically larger than the other. The Annals of Mathematical Statistics 18 (1): 50–60, 1947.

Reiss, A., Schmidt, A., Lorenz, L., Diewald, S., Hirche, S., and Urban, B. Introducing a new benchmarked dataset for physical activity monitoring. In Proceedings of the 5th International Conference on PErvasive Technologies Related to Assistive Environments (PETRA). ACM, pp. 1–8, 2012.

Reyes-Ortiz, J., Anguita, D., Ghio, A., Oneto, L., and Parra, X. Human activity recognition using smartphones. UCI Machine Learning Repository, 2012.

Zhao, B., Xiao, X., Gan, G., Zhang, B., and Xia, S. Maintaining discrimination and fairness in class incremental learning. CoRR vol. abs/1911.07053, 2019.