Kolmogorov-Arnold Networks (KANs) for Time Series Analysis
It addresses forecasting challenges in domains like satellite traffic, but is incremental as it adapts an existing KAN method to a new application.
This paper tackled time series forecasting by applying Kolmogorov-Arnold Networks (KANs) to a satellite traffic task, resulting in more accurate predictions with fewer parameters compared to Multi-Layer Perceptrons.
This paper introduces a novel application of Kolmogorov-Arnold Networks (KANs) to time series forecasting, leveraging their adaptive activation functions for enhanced predictive modeling. Inspired by the Kolmogorov-Arnold representation theorem, KANs replace traditional linear weights with spline-parametrized univariate functions, allowing them to learn activation patterns dynamically. We demonstrate that KANs outperforms conventional Multi-Layer Perceptrons (MLPs) in a real-world satellite traffic forecasting task, providing more accurate results with considerably fewer number of learnable parameters. We also provide an ablation study of KAN-specific parameters impact on performance. The proposed approach opens new avenues for adaptive forecasting models, emphasizing the potential of KANs as a powerful tool in predictive analytics.