AdaProp: Learning Adaptive Propagation for Graph Neural Network based Knowledge Graph Reasoning
This work addresses a specific bottleneck in knowledge graph reasoning for researchers and practitioners by proposing an incremental and learning-based method to enhance efficiency and relevance, though it is incremental in nature.
The paper tackles the problem of inefficient and irrelevant entity propagation in Graph Neural Network-based knowledge graph reasoning by introducing AdaProp, which learns adaptive propagation paths to filter out irrelevant entities while preserving promising targets, achieving linear complexity and improved performance in experiments.
Due to the popularity of Graph Neural Networks (GNNs), various GNN-based methods have been designed to reason on knowledge graphs (KGs). An important design component of GNN-based KG reasoning methods is called the propagation path, which contains a set of involved entities in each propagation step. Existing methods use hand-designed propagation paths, ignoring the correlation between the entities and the query relation. In addition, the number of involved entities will explosively grow at larger propagation steps. In this work, we are motivated to learn an adaptive propagation path in order to filter out irrelevant entities while preserving promising targets. First, we design an incremental sampling mechanism where the nearby targets and layer-wise connections can be preserved with linear complexity. Second, we design a learning-based sampling distribution to identify the semantically related entities. Extensive experiments show that our method is powerful, efficient, and semantic-aware. The code is available at https://github.com/LARS-research/AdaProp.