KAN-MLP-Mixer: A comprehensive investigation of the usage of Kolmogorov-Arnold Networks (KANs) for improving IMU-based Human Activity Recognition
KAN-MLP-Mixer: A comprehensive investigation of the usage of Kolmogorov-Arnold Networks (KANs) for improving IMU-based Human Activity Recognition
KAN-MLP-Mixer：关于利用柯尔莫哥洛夫-阿诺德网络（KANs）改进基于惯性测量单元（IMU）的人体活动识别的综合研究
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.
摘要： 柯尔莫哥洛夫-阿诺德网络（KANs）在学习干净、低维数据的复杂函数方面表现出了卓越的能力，但在处理嘈杂且不完美的现实世界数据集时，往往难以保持性能。相比之下，传统的多层感知机（MLPs）对噪声的容忍度更高，且计算效率更佳。在人体活动识别（HAR）模型中，将所有 MLP 组件替换为 KANs 往往会降低准确性和计算效率，这凸显了一个亟待解决的挑战：如何将 KANs 的精度与 MLPs 的噪声鲁棒性和效率相结合。
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.
为了解决这一问题，我们系统地探索了 KAN 模块在深度 HAR 网络中的各种布局，并提出了一种混合架构，旨在战略性地协同两种范式的优势。该架构使用基于 KAN 的输入嵌入层，保留用于中间特征混合的 MLP 层，并引入了一个专门的 LarctanKAN 模块用于最终的活动分类。
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.
在八个公开的 HAR 数据集上，混合 KAN-MLP 模型相较于纯 MLP 模型，平均宏观 F1 分数相对提升了 5.33%，显著优于独立的 KAN 和 MLP 基准模型。此外，将这种混合策略集成到其他最先进的 HAR 架构中，也能持续提升其性能。我们的研究结果表明，通过精心编排 KAN、MLP 或其他传统神经组件的组合，可以为现实世界的可穿戴传感环境构建更稳健、更准确的 HAR 模型。