Kolmogorov-Arnold Networks (KANs) for Time Series Analysis

TL;DR

<3-5 sentence high-level summary>

Abstract

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.

Figures (4)

• Figure 1: Example of normalized satellite traffic series data with the conditioning and prediction lengths denoted in blue, and red, respectively.
• Figure 2: Example of the flow of information in the KAN network architecture for our traffic forecasting task. Learnable activations are represented inside a square box.
• Figure 3: Satellite traffic over three different beams with their forecasted values using a 4-depth KAN and a 4-depth MLP.
• Figure 4: Ablation comparison of KAN-specific parameters during training time.