Bayesian optimization (BO) is a model-based approach for gradient-free black-box function optimization. Typically, BO is powered by a Gaussian process (GP), whose algorithmic complexity is cubic in the number of evaluations. Hence, GP-based BO cannot leverage large amounts of past or related function evaluations, for example, to warm start the BO procedure. We develop a multiple adaptive Bayesian linear regression model as a scalable alternative whose complexity is linear in the number of observations. The multiple Bayesian linear regression models are coupled through a shared feedforward neural network, which learns a joint representation and transfers knowledge across machine learning problems.
@article{arxiv.1712.02902,
title = {Multiple Adaptive Bayesian Linear Regression for Scalable Bayesian Optimization with Warm Start},
author = {Valerio Perrone and Rodolphe Jenatton and Matthias Seeger and Cedric Archambeau},
journal= {arXiv preprint arXiv:1712.02902},
year = {2017}
}