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

In three linesComparative study of Kolmogorov-Arnold Networks (KANs) vs MLPs for IMU-based Human Activity Recognition (HAR). KANs excel on clean data but fail on noisy real-world datasets. Proposed hybrid KAN-MLP architecture achieves +5.33% macro F1-score improvement across 8 public datasets, outperforming pure baselines.

Summary generated by Claude — human-verified