Bayesian Inference Made Easy via Auxiliary Augmentations

11.12.2019, 12:30-14:00  –  Campus Golm, Haus 14, Raum 0.26/27

Theo Galy-Fajou (TU Berlin)

Bayesian Inference is almost always a very challenging mathematical and computational problem. In the context of Gaussian Process, only a Gaussian likelihood leads to an analytically tractable posterior.
We extend this tractability to a large class of non-conjugate likelihoods by augmenting our model with auxiliary variables.
The complete conditionals of the model can be computed in closed-form and we show two inference schemes given analytically : variational inference and Gibbs sampling.
We show the efficiency of our approach, and how it can be automatically applied without the need of mathematical derivations.

invited by Noa Malem Shinitski

zu den Veranstaltungen