DAFOS trains GNNs faster by starting with small fanout, increasing it when loss plateaus, and prioritizing high-degree nodes, but its speedup claims are confounded by an unfair early stopping comparison.
Calibrate and Debias Layer-wise Sampling for Graph Convolutional Networks
1 Pith paper cite this work. Polarity classification is still indexing.
abstract
Multiple sampling-based methods have been developed for approximating and accelerating node embedding aggregation in graph convolutional networks (GCNs) training. Among them, a layer-wise approach recursively performs importance sampling to select neighbors jointly for existing nodes in each layer. This paper revisits the approach from a matrix approximation perspective, and identifies two issues in the existing layer-wise sampling methods: suboptimal sampling probabilities and estimation biases induced by sampling without replacement. To address these issues, we accordingly propose two remedies: a new principle for constructing sampling probabilities and an efficient debiasing algorithm. The improvements are demonstrated by extensive analyses of estimation variance and experiments on common benchmarks. Code and algorithm implementations are publicly available at https://github.com/ychen-stat-ml/GCN-layer-wise-sampling .
citation-role summary
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
REJECT 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
DAFOS: Dynamic Adaptive Fanout Optimization Sampler
DAFOS trains GNNs faster by starting with small fanout, increasing it when loss plateaus, and prioritizing high-degree nodes, but its speedup claims are confounded by an unfair early stopping comparison.