In this section
@inproceedings{Salimbeni:2017,
author = {Salimbeni, H and Deisenroth, M},
pages = {4589--4600},
publisher = {Advances in Neural Information Processing Systems (NIPS)},
title = {Doubly stochastic variational inference for deep Gaussian processes},
url = {http://hdl.handle.net/10044/1/52547},
year = {2017}
}
TY - CPAPER
AB - Gaussian processes (GPs) are a good choice for function approximation as theyare flexible, robust to over-fitting, and provide well-calibrated predictiveuncertainty. Deep Gaussian processes (DGPs) are multi-layer generalisations ofGPs, but inference in these models has proved challenging. Existing approachesto inference in DGP models assume approximate posteriors that forceindependence between the layers, and do not work well in practice. We present adoubly stochastic variational inference algorithm, which does not forceindependence between layers. With our method of inference we demonstrate that aDGP model can be used effectively on data ranging in size from hundreds to abillion points. We provide strong empirical evidence that our inference schemefor DGPs works well in practice in both classification and regression.
AU - Salimbeni,H
AU - Deisenroth,M
EP - 4600
PB - Advances in Neural Information Processing Systems (NIPS)
PY - 2017///
SN - 1049-5258
SP - 4589
TI - Doubly stochastic variational inference for deep Gaussian processes
UR - http://hdl.handle.net/10044/1/52547
ER -