KAN-GCN: Combining Kolmogorov-Arnold Network with Graph Convolution Network for an Accurate Ice Sheet Emulator
This work addresses the need for efficient ice sheet modeling for climate science applications, but it is incremental as it builds on existing GCN and KAN methods with a hybrid approach.
The paper tackled the problem of creating a fast and accurate emulator for ice sheet modeling by introducing KAN-GCN, which combines a Kolmogorov-Arnold Network with Graph Convolution Networks, resulting in improved accuracy and efficiency, with KAN-GCN matching or exceeding baseline methods across various architectures and showing better inference throughput on coarser meshes.
We introduce KAN-GCN, a fast and accurate emulator for ice sheet modeling that places a Kolmogorov-Arnold Network (KAN) as a feature-wise calibrator before graph convolution networks (GCNs). The KAN front end applies learnable one-dimensional warps and a linear mixing step, improving feature conditioning and nonlinear encoding without increasing message-passing depth. We employ this architecture to improve the performance of emulators for numerical ice sheet models. Our emulator is trained and tested using 36 melting-rate simulations with 3 mesh-size settings for Pine Island Glacier, Antarctica. Across 2- to 5-layer architectures, KAN-GCN matches or exceeds the accuracy of pure GCN and MLP-GCN baselines. Despite a small parameter overhead, KAN-GCN improves inference throughput on coarser meshes by replacing one edge-wise message-passing layer with a node-wise transform; only the finest mesh shows a modest cost. Overall, KAN-first designs offer a favorable accuracy vs. efficiency trade-off for large transient scenario sweeps.