Measuring Task Similarity and Its Implication in Fine-Tuning Graph Neural Networks

The paradigm of pre-training and fine-tuning graph neural networks has attracted wide research attention. In previous studies, the pre-trained models are viewed as universally versatile, and applied to a diverse range of downstream tasks. In many situations, however, this practice results in limited or even negative transfer. This paper, for the first time, studies the specific application scope of graph pre-trained models, i.e., the extent to which downstream tasks can benefit from specific pre-training tasks. We find that not all downstream tasks can effectively benefit from a graph pre-trained model. In light of this, we introduce the measure task consistency to quantify the similarity between graph pre-training and downstream tasks. This measure assesses the extent to which downstream tasks can benefit from specific pre-training tasks. Moreover, a novel fine-tuning strategy, Bridge-Tune, is proposed to further diminish the impact of the difference between pre-training and downstream tasks. The key innovation in Bridge-Tune is an intermediate step that bridges pre-training and downstream tasks. This step takes into account the task differences and further refines the pre-trained model. The superiority of the presented fine-tuning strategy is validated via numerous experiments with different pre-trained models and downstream tasks.