Evaluation of Regularization-based Continual Learning Approaches: Application to HAR

TL;DR

This paper compares three regularization-based continual learning methods applied to Human Activity Recognition, highlighting their strengths and limitations through experiments on the UCI HAR dataset.

Contribution

It provides the first comprehensive comparison of these methods in the HAR domain, offering insights into their relative performance and applicability.

Findings

No single method outperforms others in all scenarios

Regularization techniques show promise but have limitations

Experimental results vary depending on the scenario

Abstract

Pervasive computing allows the provision of services in many important areas, including the relevant and dynamic field of health and well-being. In this domain, Human Activity Recognition (HAR) has gained a lot of attention in recent years. Current solutions rely on Machine Learning (ML) models and achieve impressive results. However, the evolution of these models remains difficult, as long as a complete retraining is not performed. To overcome this problem, the concept of Continual Learning is very promising today and, more particularly, the techniques based on regularization. These techniques are particularly interesting for their simplicity and their low cost. Initial studies have been conducted and have shown promising outcomes. However, they remain very specific and difficult to compare. In this paper, we provide a comprehensive comparison of three regularization-based methods that we adapted to the HAR domain, highlighting their strengths and limitations. Our experiments were conducted on the UCI HAR dataset and the results showed that no single technique outperformed all others in all scenarios considered.