Unleash Graph Neural Networks from Heavy Tuning
This addresses the challenge of efficient GNN deployment for researchers and practitioners by reducing tuning burdens, though it is incremental as it builds on existing diffusion methods.
The paper tackles the problem of high computational cost and human effort in hyperparameter tuning for Graph Neural Networks (GNNs) by proposing a graph conditional latent diffusion framework (GNN-Diff) that generates high-performing GNNs directly from light-tuning checkpoints, outperforming comprehensive grid search.
Graph Neural Networks (GNNs) are deep-learning architectures designed for graph-type data, where understanding relationships among individual observations is crucial. However, achieving promising GNN performance, especially on unseen data, requires comprehensive hyperparameter tuning and meticulous training. Unfortunately, these processes come with high computational costs and significant human effort. Additionally, conventional searching algorithms such as grid search may result in overfitting on validation data, diminishing generalization accuracy. To tackle these challenges, we propose a graph conditional latent diffusion framework (GNN-Diff) to generate high-performing GNNs directly by learning from checkpoints saved during a light-tuning coarse search. Our method: (1) unleashes GNN training from heavy tuning and complex search space design; (2) produces GNN parameters that outperform those obtained through comprehensive grid search; and (3) establishes higher-quality generation for GNNs compared to diffusion frameworks designed for general neural networks.