Timezone: »

Decoupling the Depth and Scope of Graph Neural Networks
Hanqing Zeng · Muhan Zhang · Yinglong Xia · Ajitesh Srivastava · Andrey Malevich · Rajgopal Kannan · Viktor Prasanna · Long Jin · Ren Chen

Thu Dec 09 04:30 PM -- 06:00 PM (PST) @ Virtual #None

State-of-the-art Graph Neural Networks (GNNs) have limited scalability with respect to the graph and model sizes. On large graphs, increasing the model depth often means exponential expansion of the scope (i.e., receptive field). Beyond just a few layers, two fundamental challenges emerge: 1. degraded expressivity due to oversmoothing, and 2. expensive computation due to neighborhood explosion. We propose a design principle to decouple the depth and scope of GNNs – to generate representation of a target entity (i.e., a node or an edge), we first extract a localized subgraph as the bounded-size scope, and then apply a GNN of arbitrary depth on top of the subgraph. A properly extracted subgraph consists of a small number of critical neighbors, while excluding irrelevant ones. The GNN, no matter how deep it is, smooths the local neighborhood into informative representation rather than oversmoothing the global graph into “white noise”. Theoretically, decoupling improves the GNN expressive power from the perspectives of graph signal processing (GCN), function approximation (GraphSAGE) and topological learning (GIN). Empirically, on seven graphs (with up to 110M nodes) and six backbone GNN architectures, our design achieves significant accuracy improvement with orders of magnitude reduction in computation and hardware cost.

Author Information

Hanqing Zeng (University of Southern California)
Muhan Zhang (Peking University)
Yinglong Xia (University of Southern California)
Ajitesh Srivastava (University of Southern California)
Andrey Malevich (Facebook)
Rajgopal Kannan (DoD HPCMP)
Viktor Prasanna (University of Southern California)
Long Jin (Facebook)
Ren Chen (University of Southern California)

More from the Same Authors