Doubly Stochastic Variational Inference for Deep Gaussian Processes
This addresses the practical problem of scalable and accurate inference in DGPs for researchers and practitioners in machine learning, representing a novel method rather than an incremental improvement.
The authors tackled the challenge of inference in deep Gaussian processes (DGPs) by developing a doubly stochastic variational inference algorithm that avoids forcing independence between layers, enabling effective use on datasets ranging from hundreds to a billion points with strong empirical performance in classification and regression.
Gaussian processes (GPs) are a good choice for function approximation as they are flexible, robust to over-fitting, and provide well-calibrated predictive uncertainty. Deep Gaussian processes (DGPs) are multi-layer generalisations of GPs, but inference in these models has proved challenging. Existing approaches to inference in DGP models assume approximate posteriors that force independence between the layers, and do not work well in practice. We present a doubly stochastic variational inference algorithm, which does not force independence between layers. With our method of inference we demonstrate that a DGP model can be used effectively on data ranging in size from hundreds to a billion points. We provide strong empirical evidence that our inference scheme for DGPs works well in practice in both classification and regression.