Inductive Graph Alignment Prompt: Bridging the Gap between Graph Pre-training and Inductive Fine-tuning From Spectral Perspective
This work addresses the problem of generalizing graph pre-training to inductive scenarios for researchers and practitioners in graph machine learning, offering a novel approach to bridge data gaps, though it is incremental as it builds on existing prompt-based methods.
The paper tackles the performance limitation in graph neural networks due to data and task gaps between pre-training and fine-tuning stages, proposing the Inductive Graph Alignment Prompt (IGAP) method that bridges graph signal and structure gaps in spectral space, achieving successful results in node and graph classification tasks across transductive, semi-inductive, and inductive settings.
The "Graph pre-training and fine-tuning" paradigm has significantly improved Graph Neural Networks(GNNs) by capturing general knowledge without manual annotations for downstream tasks. However, due to the immense gap of data and tasks between the pre-training and fine-tuning stages, the model performance is still limited. Inspired by prompt fine-tuning in Natural Language Processing(NLP), many endeavors have been made to bridge the gap in graph domain. But existing methods simply reformulate the form of fine-tuning tasks to the pre-training ones. With the premise that the pre-training graphs are compatible with the fine-tuning ones, these methods typically operate in transductive setting. In order to generalize graph pre-training to inductive scenario where the fine-tuning graphs might significantly differ from pre-training ones, we propose a novel graph prompt based method called Inductive Graph Alignment Prompt(IGAP). Firstly, we unify the mainstream graph pre-training frameworks and analyze the essence of graph pre-training from graph spectral theory. Then we identify the two sources of the data gap in inductive setting: (i) graph signal gap and (ii) graph structure gap. Based on the insight of graph pre-training, we propose to bridge the graph signal gap and the graph structure gap with learnable prompts in the spectral space. A theoretical analysis ensures the effectiveness of our method. At last, we conduct extensive experiments among nodes classification and graph classification tasks under the transductive, semi-inductive and inductive settings. The results demonstrate that our proposed method can successfully bridge the data gap under different settings.