Community-structured sparse relational graphs improve 5-layer CIFAR-10 accuracy over fully connected baselines, but the advantage reverses at 8 layers due to over-smoothing.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Effects of relational graph modularity and depth on the learning performance of neural networks
Community-structured sparse relational graphs improve 5-layer CIFAR-10 accuracy over fully connected baselines, but the advantage reverses at 8 layers due to over-smoothing.