论文标题
关于图形神经网络的预测不稳定
On the Prediction Instability of Graph Neural Networks
论文作者
论文摘要
受过训练的模型的不稳定性，即单个节点预测对随机因素的依赖性会影响可重复性，可靠性和对机器学习系统的信任。在本文中，我们通过最先进的图形神经网络（GNNS）系统地评估了节点分类的预测不稳定性。通过我们的实验，我们确定了经过相同模型超参数的相同数据训练的流行GNN模型的多次实例化导致了几乎相同的汇总性能，但在单个节点的预测中表现出了很大的分歧。我们发现，在算法运行中，最多三分之一的错误分类节点不同。我们确定超参数，节点属性和训练设置的大小与预测的稳定性之间的相关性。通常，最大化模型性能也隐含地减少了模型不稳定性。
Instability of trained models, i.e., the dependence of individual node predictions on random factors, can affect reproducibility, reliability, and trust in machine learning systems. In this paper, we systematically assess the prediction instability of node classification with state-of-the-art Graph Neural Networks (GNNs). With our experiments, we establish that multiple instantiations of popular GNN models trained on the same data with the same model hyperparameters result in almost identical aggregated performance but display substantial disagreement in the predictions for individual nodes. We find that up to one third of the incorrectly classified nodes differ across algorithm runs. We identify correlations between hyperparameters, node properties, and the size of the training set with the stability of predictions. In general, maximizing model performance implicitly also reduces model instability.
