KAN-HAR: A Human activity recognition based on Kolmogorov-Arnold Network

TL;DR

This paper introduces KAN-HAR, a novel human activity recognition method using Kolmogorov-Arnold Networks with a single accelerometer, achieving high accuracy with fewer parameters and better interpretability than traditional deep learning models.

Contribution

The work presents a new application of Kolmogorov-Arnold Networks for HAR, demonstrating improved efficiency and interpretability over conventional deep neural networks.

Findings

KAN outperforms traditional DNNs in accuracy on MotionSense dataset.

KAN requires significantly fewer parameters, enhancing efficiency.

The approach offers better interpretability for HAR models.

Abstract

Human Activity Recognition (HAR) plays a critical role in numerous applications, including healthcare monitoring, fitness tracking, and smart environments. Traditional deep learning (DL) approaches, while effective, often require extensive parameter tuning and may lack interpretability. In this work, we investigate the use of a single three-axis accelerometer and the Kolmogorov--Arnold Network (KAN) for HAR tasks, leveraging its ability to model complex nonlinear relationships with improved interpretability and parameter efficiency. The MotionSense dataset, containing smartphone-based motion sensor signals across various physical activities, is employed to evaluate the proposed approach. Our methodology involves preprocessing and normalization of accelerometer and gyroscope data, followed by KAN-based feature learning and classification. Experimental results demonstrate that the KAN achieves competitive or superior classification performance compared to conventional deep neural networks, while maintaining a significantly reduced parameter count. This highlights the potential of KAN architectures as an efficient and interpretable alternative for real-world HAR systems. The open-source implementation of the proposed framework is available at the Project's GitHub Repository.