2 papers
cs.LG2026
MQ-GNN: A Multi-Queue Pipelined Architecture for Scalable and Efficient GNN Training
Irfan Ullah, Young-Koo Lee
Graph Neural Networks (GNNs) are powerful tools for learning graph-structured data, but their scalability is hindered by inefficient mini-batch generation, data transfer bottleneck…
cs.LG2025
DAFOS: Dynamic Adaptive Fanout Optimization Sampler
Irfan Ullah, Young-Koo Lee
Graph Neural Networks (GNNs) are becoming an essential tool for learning from graph-structured data, however uniform neighbor sampling and static fanout settings frequently limit G…