KAN-MLP-Mixer: A comprehensive investigation of the usage of Kolmogorov-Arnold Networks (KANs) for improving IMU-based Human Activity Recognition

TL;DR

This paper introduces a hybrid KAN-MLP architecture that combines Kolmogorov-Arnold Networks and traditional MLPs to enhance accuracy and robustness in IMU-based Human Activity Recognition across diverse datasets.

Contribution

It systematically explores KAN placement in HAR models and proposes a hybrid architecture that leverages the strengths of both KANs and MLPs, achieving significant performance improvements.

Findings

Hybrid KAN-MLP model improves macro F1 score by 5.33% over pure-MLP.

The hybrid approach outperforms standalone KAN and MLP baselines.

Integrating the hybrid strategy into other architectures boosts their performance.

Abstract

Kolmogorov-Arnold Networks (KANs) have demonstrated an exceptional ability to learn complex functions on clean, low-dimensional data but struggle to maintain performance on noisy and imperfect real-world datasets. In contrast, conventional multi-layer perceptrons (MLPs) are far more tolerant to noise and computationally efficient. Replacing all MLP components with KANs in HAR models often degrades accuracy and computation efficiency, highlighting an open challenge: how to combine KANs' precision with MLPs' noise robustness and efficiency. To address this, we systematically explore various placements of KAN modules within deep HAR networks and propose a hybrid architecture that strategically synergizes the strengths of both paradigms, which uses a KAN-based input embedding layer, retains MLP layers for intermediate feature mixing, and introduces a specialized LarctanKAN module for final activity classification. Across eight public HAR datasets, the hybrid KAN-MLP model achieves an average macro F1 score relative improvement of 5.33\% compared pure-MLP model, significantly outperforming standalone KAN and MLP baselines. Furthermore, integrating this hybrid strategy into other state-of-the-art HAR architectures consistently boosts their performance. Our findings demonstrate that a carefully orchestrated combination of KAN, MLP, or other conventional neural components yields more robust and accurate HAR models for real-world wearable sensing environments.