An information-theoretic framework generates novel graphs via latent mixture modeling with novelty and reliability enforced through minimum description length, backed by convergence guarantees and experiments on graph datasets.
Unsupervised meta-learning through latent-space interpolation in generative models.arXiv preprint arXiv:2006.10236, 2020
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
An Information Theoretic Framework for Graph Novelty Generation via Latent Mixture Modeling
An information-theoretic framework generates novel graphs via latent mixture modeling with novelty and reliability enforced through minimum description length, backed by convergence guarantees and experiments on graph datasets.