Graph2Graph Learning with Conditional Autoregressive Models
This addresses the need for more structured representations in graph learning tasks like generative modeling, which is incremental as it builds on existing graph neural network methods.
The paper tackles the problem of graph-to-graph learning, where models must predict graph-structured outputs, by introducing a conditional autoregressive model. It demonstrates results on subgraph prediction, graph autoencoding, and pretraining for graph classification, though no concrete numbers are provided in the abstract.
We present a graph neural network model for solving graph-to-graph learning problems. Most deep learning on graphs considers ``simple'' problems such as graph classification or regressing real-valued graph properties. For such tasks, the main requirement for intermediate representations of the data is to maintain the structure needed for output, i.e., keeping classes separated or maintaining the order indicated by the regressor. However, a number of learning tasks, such as regressing graph-valued output, generative models, or graph autoencoders, aim to predict a graph-structured output. In order to successfully do this, the learned representations need to preserve far more structure. We present a conditional auto-regressive model for graph-to-graph learning and illustrate its representational capabilities via experiments on challenging subgraph predictions from graph algorithmics; as a graph autoencoder for reconstruction and visualization; and on pretraining representations that allow graph classification with limited labeled data.