Bayesian Quadrature for Prediction and Optimisation
Add to your list(s)
Download to your calendar using vCal
If you have a question about this talk, please contact Zoubin Ghahramani.
Bayesian inference often requires the evaluation of nonanalytic integrals. These must be approximated using methods for numerical integration, or quadrature, of which Monte Carlo techniques are the most common. A more principled alternative is found in Bayesian Quadrature, also known as Bayesian Monte Carlo, which employs a Gaussian process model for the integrand. We describe an extension of previous Bayesian quadrature techniques that explicitly models the non-negativity of integrands. We then present results of the use of Bayesian quadrature for problems related to changepoint and fault detection, global optimisation and sensor selection.
This talk is part of the Machine Learning @ CUED series.
This talk is included in these lists:
Note that ex-directory lists are not shown.
|