High-Order Pooling for Graph Neural Networks with Tensor Decomposition
This work addresses the problem of limited expressiveness in GNNs for researchers and practitioners, offering a novel method that is incremental in improving pooling mechanisms.
The authors tackled the limitation of simple linear pooling operations in Graph Neural Networks (GNNs) by proposing the Tensorized Graph Neural Network (tGNN), which uses tensor decomposition to model high-order non-linear node interactions, achieving state-of-the-art results on OGB datasets for node and graph classification.
Graph Neural Networks (GNNs) are attracting growing attention due to their effectiveness and flexibility in modeling a variety of graph-structured data. Exiting GNN architectures usually adopt simple pooling operations (eg. sum, average, max) when aggregating messages from a local neighborhood for updating node representation or pooling node representations from the entire graph to compute the graph representation. Though simple and effective, these linear operations do not model high-order non-linear interactions among nodes. We propose the Tensorized Graph Neural Network (tGNN), a highly expressive GNN architecture relying on tensor decomposition to model high-order non-linear node interactions. tGNN leverages the symmetric CP decomposition to efficiently parameterize permutation-invariant multilinear maps for modeling node interactions. Theoretical and empirical analysis on both node and graph classification tasks show the superiority of tGNN over competitive baselines. In particular, tGNN achieves the most solid results on two OGB node classification datasets and one OGB graph classification dataset.