Recyclable Gaussian Processes
This addresses the computational bottleneck of reusing multiple Gaussian process approximations in distributed settings, though it appears incremental as it builds on existing variational methods.
The paper tackles the problem of efficiently combining multiple pre-trained Gaussian process models without retraining on original data, enabling regression, classification, and heterogeneous tasks. The results demonstrate the framework's usability in large-scale distributed experiments with comparisons to exact inference models.
We present a new framework for recycling independent variational approximations to Gaussian processes. The main contribution is the construction of variational ensembles given a dictionary of fitted Gaussian processes without revisiting any subset of observations. Our framework allows for regression, classification and heterogeneous tasks, i.e. mix of continuous and discrete variables over the same input domain. We exploit infinite-dimensional integral operators based on the Kullback-Leibler divergence between stochastic processes to re-combine arbitrary amounts of variational sparse approximations with different complexity, likelihood model and location of the pseudo-inputs. Extensive results illustrate the usability of our framework in large-scale distributed experiments, also compared with the exact inference models in the literature.