Localized Weather Prediction Using Kolmogorov-Arnold Network-Based Models and Deep RNNs

Code

• github.com/AngeClementAkazan/Localized-Weather-Prediction-Using-KAN-and-DeepRNNs
  OfficialIn paperpytorch★ 6

Abstract

Weather forecasting is crucial for managing risks and economic planning, particularly in tropical Africa, where extreme events severely impact livelihoods. Yet, existing forecasting methods often struggle with the region's complex, non-linear weather patterns. This study benchmarks deep recurrent neural networks such as LSTM, GRU, BiLSTM, BiGRU, and Kolmogorov-Arnold-based models (KAN and TKAN) for daily forecasting of temperature, precipitation, and pressure in two tropical cities: Abidjan, Cote d'Ivoire (Ivory Coast) and Kigali (Rwanda). We further introduce two customized variants of TKAN that replace its original SiLU activation function with GeLU and MiSH, respectively. Using station-level meteorological data spanning from 2010 to 2024, we evaluate all the models on standard regression metrics. KAN achieves temperature prediction (R^2=0.9986 in Abidjan, 0.9998 in Kigali, MSE < 0.0014~^ C ^2), while TKAN variants minimize absolute errors for precipitation forecasting in low-rainfall regimes. The customized TKAN models demonstrate improvements over the standard TKAN across both datasets. Classical RNNs remain highly competitive for atmospheric pressure (R^2 0.83-0.86), outperforming KAN-based models in this task. These results highlight the potential of spline-based neural architectures for efficient and data-efficient forecasting.

Tasks

Reproductions