Timezone: »

Modulating Surrogates for Bayesian Optimization
Erik Bodin · Markus Kaiser · Ieva Kazlauskaite · Zhenwen Dai · Neill Campbell · Carl Henrik Ek

Thu Jul 16 01:00 PM -- 01:45 PM & Fri Jul 17 02:00 AM -- 02:45 AM (PDT) @ None #None

Bayesian optimization (BO) methods often rely on the assumption that the objective function is well-behaved, but in practice, this is seldom true for real-world objectives even if noise-free observations can be collected. Common approaches, which try to model the objective as precisely as possible, often fail to make progress by spending too many evaluations modeling irrelevant details. We address this issue by proposing surrogate models that focus on the well-behaved structure in the objective function, which is informative for search, while ignoring detrimental structure that is challenging to model from few observations. First, we demonstrate that surrogate models with appropriate noise distributions can absorb challenging structures in the objective function by treating them as irreducible uncertainty. Secondly, we show that a latent Gaussian process is an excellent surrogate for this purpose, comparing with Gaussian processes with standard noise distributions. We perform numerous experiments on a range of BO benchmarks and find that our approach improves reliability and performance when faced with challenging objective functions.

Author Information

Erik Bodin (University of Bristol)
Markus Kaiser (Technical University Munich)
Ieva Kazlauskaite (University of Bath)
Zhenwen Dai (Spotify)
Neill Campbell (University of Bath)
Carl Henrik Ek (University of Cambridge)

More from the Same Authors