Variational inference based on a subclass of closed skew normals
LSL Tan, A Chen - Journal of Computational and Graphical …, 2024 - Taylor & Francis
Gaussian distributions are widely used in Bayesian variational inference to approximate
intractable posterior densities, but the ability to accommodate skewness can improve …
intractable posterior densities, but the ability to accommodate skewness can improve …
Skew-symmetric approximations of posterior distributions
Routinely-implemented deterministic approximations of posterior distributions from, eg,
Laplace method, variational Bayes and expectation-propagation, generally rely on …
Laplace method, variational Bayes and expectation-propagation, generally rely on …