pith. sign in

arxiv: 1812.06067 · v1 · pith:HILZAX2Knew · submitted 2018-12-14 · 📊 stat.ML · cs.LG

Non-Factorised Variational Inference in Dynamical Systems

classification 📊 stat.ML cs.LG
keywords functiontransitiondynamicalinferenceposteriorprocesssystemsystems
0
0 comments X p. Extension
pith:HILZAX2K Add to your LaTeX paper What is a Pith Number?
\usepackage{pith}
\pithnumber{HILZAX2K}

Prints a linked pith:HILZAX2K badge after your title and writes the identifier into PDF metadata. Compiles on arXiv with no extra files. Learn more

read the original abstract

We focus on variational inference in dynamical systems where the discrete time transition function (or evolution rule) is modelled by a Gaussian process. The dominant approach so far has been to use a factorised posterior distribution, decoupling the transition function from the system states. This is not exact in general and can lead to an overconfident posterior over the transition function as well as an overestimation of the intrinsic stochasticity of the system (process noise). We propose a new method that addresses these issues and incurs no additional computational costs.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.