analiza aiqbench
This study addresses the trade-off between Kolmogorov-Arnold Networks' precision and Multi-Layer Perceptrons' noise robustness by proposing a hybrid architecture for IMU-based Human Activity Recognition. By strategically placing KAN modules at the input and output layers while retaining MLPs for intermediate processing, the model achieves a 5.33% relative improvement in macro F1 score across eight datasets. The findings suggest that selectively combining these paradigms yields more accurate and efficient models for real-world wearable sensing applications.
KAN-MLP-Mixer: A comprehensive investigation of the usage of Kolmogorov-Arnold Networks (KANs) for improving IMU-based Human Activity Recognition
arXiv:2605.19031v2 Announce Type: replace 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.
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.
Artificial Intelligence (cs.AI); Signal Processing (eess.SP)
arXiv:2605.19031 [cs.AI]
(or arXiv:2605.19031v2 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2605.19031 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 290 cuvinte
Subiecte conexe
SW-KAN: Kolmogorov-Arnold Networks with Stieltjes-Wigert q-Orthogonal Polynomials
arXiv · 2 oct. 2026
KANEx: Translating Kolmogorov-Arnold Networks' Interpretability to Medical Explainability
arXiv · 28 iul. 2026
Neural scaling laws and evolution of learnable activation functions of Kolmogorov-Arnold networks
arXiv · 2 oct. 2026
Evaluating the Effectiveness of SechKAN on 1D Data
arXiv · 24 sept. 2026