A contrastive GNN framework learns transferable node embeddings for approximate k-coloring that align same-color nodes and separate adjacent ones, with analysis showing line-prototype geometry at optima and experiments showing generalization on synthetic and real graphs.
Nonsmooth optimization tech- niques on riemannian manifolds.Journal of Optimization Theory and Applications, 158(2): 328–342, 2013
1 Pith paper cite this work, alongside 21 external citations. Polarity classification is still indexing.
1
Pith paper citing it
21
external citations · OpenAlex
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Contrastive Neural Algorithmic Reasoning for Graph Coloring
A contrastive GNN framework learns transferable node embeddings for approximate k-coloring that align same-color nodes and separate adjacent ones, with analysis showing line-prototype geometry at optima and experiments showing generalization on synthetic and real graphs.