linearizedGPhttp://mloss.orgUpdates and additions to linearizedGPenFri, 28 Nov 2014 07:02:54 -0000linearizedGP 1.0<html><p>This code contains python 2.7 and 3.x implementations of the Extended and Unscented Gaussian Processes (E/UGPs) for solving inverse problems. These are similar to regular GPs, but the latent function, f, can optionally have an extra nonlinear relationship to the observations, y, in the likelihood, </p> <ul> <li> Normal GP likelihood: y ~ N(f, s^2 I_N) or for a single observation, n, y_n = f_n + e. </li> <li> E/UGP likelihood: y ~ N(g(f), s^2 I_N) or for a single observation, n, y_n = g(f_n) + e. </li> </ul> <p>Where g(.) is an arbitrary scalar function (maps R to R). The posterior Gaussian process parameters are learned using a variational objective with Gauss-Newton style linearization (of g) and mean finding. </p> <p>More information can be found in our NIPS 2014 paper here: </p> <p> </p></html>daniel steinbergFri, 28 Nov 2014 07:02:54 -0000 inferencenonparametric bayesvariational inferenceinverse methodsgaussian processbayesian inferenceblack box optimizationgeneralized regressionderivative free