Multi-Information Source Optimization

By Matthias Poloczek, Jialei Wang, Peter I. Frazier. 2016

230 citations recorded by the source corpus.

We consider Bayesian optimization of an expensive-to-evaluate black-box objective function, where we also have access to cheaper approximations of the objective. In general, such approximations arise in applications such as reinforcement learning, engineering, and the natural sciences, and are subject to an inherent, unknown bias. This model discrepancy is caused by an inadequate internal model that deviates from reality and can vary over the domain, making the utilization of these approximations a non-trivial task. We present a novel algorithm that provides a rigorous mathematical treatment of the uncertainties arising from model discrepancies and noisy observations. Its optimization decisions rely on a value of information analysis that extends the Knowledge Gradient factor to the setting of multiple information sources that vary in cost: each sampling decision maximizes the predicted benefit per unit cost. We conduct an experimental evaluation that demonstrates that the method consistently outperforms other state-of-the-art techniques: it finds designs of considerably higher objective value and additionally inflicts less cost in the exploration process.

Subfields: Artificial Intelligence, Computational Theory and Mathematics, Management Science and Operations Research

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computational Theory and Mathematics

In Management Science and Operations Research

Citations in the corpus

BOHB: Robust and Efficient Hyperparameter Optimization at Scale (2018, methodology): “…algorithms, recent methods in Bayesian optimization try to extend the traditional blackbox setting by exploiting cheaper ﬁdelities of the objective function (Swersky et al., 2014; Klein et al., 2017a; Swersky et al., 2013; Kan-dasamy et al., 2017; Klein et al., 2017c; Poloczek et al., 2017).”
Multi-fidelity Bayesian Optimisation with Continuous Approximations (2017, methodology): “BO techniques have been used in developing multi-fidelity optimisation methods in various applications such as hyperparameter tuning and industrial design (Forrester et al., 2007; Huang et al., 2006; Klein et al., 2015; Lam et al., 2015; Poloczek et al., 2016; Swersky et al., 2013).”
A Tutorial on Bayesian Optimization (2018, background): “When such phenomenon are first-order, KG tends to significantly outperform EI (Wu et al., 2017; Poloczek et al., 2017; Wu and Frazier, 2016; Toscano-Palmerin and Frazier, 2018).”
BoTorch: Programmable Bayesian Optimization in PyTorch (2019, methodology): “For simplicity, we only consider the standard BO problem, but multi-ﬁdelity extensions (Poloczek et al., 2017; Wu et al., 2019) can also easily be used.”
Bayesian Optimization is Superior to Random Search for Machine Learning Hyperparameter Tuning: Analysis of the Black-Box Optimization Challenge 2020 (2021, background): “Multi-ﬁdelity (or multi-information source) computations were not available in this competition, but they may be common in practical circumstances [53].”
Multi-Fidelity Black-Box Optimization with Hierarchical Partitions (2018, background): “Multi-fidelity optimization has had a rich history in many settings (Agarwal et al., 2011; Forrester et al., 2007; Huang et al., 2006a; Klein et al., 2016; Lam et al., 2015; Li et al., 2016; Poloczek et al., 2016; Sabharwal et al., 2015; Zhang & Chaudhuri, 2015), with those that are not applications”