Scaling Knowledge Graph Embedding Models
This addresses computational bottlenecks for researchers and practitioners working with large-scale knowledge graphs.
The paper tackles the challenge of scaling Graph Neural Network training for knowledge graph link prediction by proposing algorithmic strategies including self-sufficient partitions, constraint-based negative sampling, and edge mini-batch training, achieving a 16x speedup on benchmark datasets while maintaining comparable model performance to non-distributed methods.
Developing scalable solutions for training Graph Neural Networks (GNNs) for link prediction tasks is challenging due to the high data dependencies which entail high computational cost and huge memory footprint. We propose a new method for scaling training of knowledge graph embedding models for link prediction to address these challenges. Towards this end, we propose the following algorithmic strategies: self-sufficient partitions, constraint-based negative sampling, and edge mini-batch training. Both, partitioning strategy and constraint-based negative sampling, avoid cross partition data transfer during training. In our experimental evaluation, we show that our scaling solution for GNN-based knowledge graph embedding models achieves a 16x speed up on benchmark datasets while maintaining a comparable model performance as non-distributed methods on standard metrics.