For strongly concave log-likelihoods, square-root (Cholesky) parametrization of Gaussian variational inference yields exponential convergence guarantees for both the natural-gradient flow and a discrete-time natural-gradient algorithm.
On the properties of variational approximations of G ibbs posteriors
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Optimization Guarantees for Square-Root Natural-Gradient Variational Inference
For strongly concave log-likelihoods, square-root (Cholesky) parametrization of Gaussian variational inference yields exponential convergence guarantees for both the natural-gradient flow and a discrete-time natural-gradient algorithm.