no code implementations • 15 Jul 2021 • Bahram Jafrasteh, Carlos Villacampa-Calvo, Daniel Hernández-Lobato
For this, we use a neural network that receives the observed data as an input and outputs the inducing points locations and the parameters of $q$.
1 code implementation • 28 Jan 2020 • Carlos Villacampa-Calvo, Bryan Zaldivar, Eduardo C. Garrido-Merchán, Daniel Hernández-Lobato
The results obtained show that, although the classification error is similar across methods, the predictive distribution of the proposed methods is better, in terms of the test log-likelihood, than the predictive distribution of a classifier based on GPs that ignores input noise.
no code implementations • ICML 2017 • Carlos Villacampa-Calvo, Daniel Hernández-Lobato
Furthermore, extra assumptions in the approximate inference process make the memory cost independent of $N$.