Models and Benchmarks for Representation Learning of Partially Observed Subgraphs

TL;DR

This paper introduces a new framework called PSI for learning representations of partially observed subgraphs, improving over existing methods by maximizing mutual information and reconstructing full subgraph representations.

Contribution

The paper formulates a novel task for partially observed subgraph representation learning and generalizes existing InfoMax models within the PSI framework, including a two-stage $k$-hop approach.

Findings

PSI models outperform baselines on real-world datasets.

The two-stage $k$-hop PSI enhances expressiveness of subgraph representations.

Mutual information maximization improves subgraph embedding quality.

Abstract

Subgraphs are rich substructures in graphs, and their nodes and edges can be partially observed in real-world tasks. Under partial observation, existing node- or subgraph-level message-passing produces suboptimal representations. In this paper, we formulate a novel task of learning representations of partially observed subgraphs. To solve this problem, we propose Partial Subgraph InfoMax (PSI) framework and generalize existing InfoMax models, including DGI, InfoGraph, MVGRL, and GraphCL, into our framework. These models maximize the mutual information between the partial subgraph's summary and various substructures from nodes to full subgraphs. In addition, we suggest a novel two-stage model with $k$-hop PSI, which reconstructs the representation of the full subgraph and improves its expressiveness from different local-global structures. Under training and evaluation protocols designed for this problem, we conduct experiments on three real-world datasets and demonstrate that PSI models outperform baselines.