LGDec 16, 2024

Asymmetric Learning for Spectral Graph Neural Networks

arXiv:2412.11739v11 citationsh-index: 3Has CodeAAAI
Originality Incremental advance
AI Analysis

This work addresses optimization difficulties in spectral GNNs, which is an incremental advancement for researchers and practitioners in graph machine learning.

The paper tackled the challenge of optimizing spectral graph neural networks (GNNs) by analyzing differences in parameters that lead to poorly conditioned problems, and introduced an asymmetric learning approach that dynamically preconditioned gradients to improve performance. Extensive experiments on eighteen benchmark datasets showed consistent improvements, especially for heterophilic graphs.

Optimizing spectral graph neural networks (GNNs) remains a critical challenge in the field, yet the underlying processes are not well understood. In this paper, we investigate the inherent differences between graph convolution parameters and feature transformation parameters in spectral GNNs and their impact on the optimization landscape. Our analysis reveals that these differences contribute to a poorly conditioned problem, resulting in suboptimal performance. To address this issue, we introduce the concept of the block condition number of the Hessian matrix, which characterizes the difficulty of poorly conditioned problems in spectral GNN optimization. We then propose an asymmetric learning approach, dynamically preconditioning gradients during training to alleviate poorly conditioned problems. Theoretically, we demonstrate that asymmetric learning can reduce block condition numbers, facilitating easier optimization. Extensive experiments on eighteen benchmark datasets show that asymmetric learning consistently improves the performance of spectral GNNs for both heterophilic and homophilic graphs. This improvement is especially notable for heterophilic graphs, where the optimization process is generally more complex than for homophilic graphs. Code is available at https://github.com/Mia-321/asym-opt.git.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes