A tangent approximation variational inference method for strongly super-Gaussian likelihoods that induces conjugacy via convex duality and provides convergence guarantees plus near-minimax risk bounds.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ME 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
A Generalized Tangent Approximation based Variational Inference Framework for Strongly Super-Gaussian Likelihoods
A tangent approximation variational inference method for strongly super-Gaussian likelihoods that induces conjugacy via convex duality and provides convergence guarantees plus near-minimax risk bounds.