A kernelized path gradient estimator with an importance-sampling correction reduces gradient variance in semi-implicit variational inference and matches or improves on KSIVI.
Learning to draw samples with amortized stein variational gradient descent
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Semi-Implicit Variational Inference via Kernelized Path Gradient Descent
A kernelized path gradient estimator with an importance-sampling correction reduces gradient variance in semi-implicit variational inference and matches or improves on KSIVI.