1 code implementation • 17 Jun 2019 • Alexander Immer, Guillaume P. Dehaene
The Black Box Variational Inference (Ranganath et al. (2014)) algorithm provides a universal method for Variational Inference, but taking advantage of special properties of the approximation family or of the target can improve the convergence speed significantly.
no code implementations • 4 Apr 2019 • Guillaume P. Dehaene
Bernstein-von Mises results (BvM) establish that the Laplace approximation is asymptotically correct in the large-data limit.
no code implementations • 24 Nov 2017 • Guillaume P. Dehaene
In this work, we present a theorem upper-bounding the KL divergence between a log-concave target density $f\left(\boldsymbol{\theta}\right)$ and its Laplace approximation $g\left(\boldsymbol{\theta}\right)$.
no code implementations • 15 Dec 2016 • Guillaume P. Dehaene
More precisely, we link it to using gradient descent to compute the Laplace approximation of a target probability distribution.