The paper introduces Semantic Variational Bayes (SVB) derived from the author's Semantic Information G Theory, claiming simpler computation than standard VB for latent variable inference via maximum G/R criterion.
Lu, A Generalized Information Theory (Chinese: 广义信息论), Hefei, China: China Science and Technology University Press(中国科学技术大学出版）, 1993
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Semantic Variational Bayes Based on Semantic Information G Theory for Solving Latent Variables
The paper introduces Semantic Variational Bayes (SVB) derived from the author's Semantic Information G Theory, claiming simpler computation than standard VB for latent variable inference via maximum G/R criterion.