Replication or Exploration? Sequential Design for Stochastic Simulation Experiments

We investigate the merits of replication, and provide methods for optimal design (including replicates), with the goal of obtaining globally accurate emulation of noisy computer simulation experiments. We first show that replication can be beneficial from both design and computational perspectives,...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Technometrics Jg. 61; H. 1; S. 7 - 23
Hauptverfasser: Binois, Mickaël, Huang, Jiangeng, Gramacy, Robert B., Ludkovski, Mike
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Alexandria Taylor & Francis 02.01.2019
American Society for Quality and the American Statistical Association
American Society for Quality
Schlagworte:
Computer experiment
Computer simulation
Data simulation
Design optimization
Epidemiology
Experiments
Gaussian process
Input-dependent noise
Inventory management
Lookahead
Normal distribution
Replicated observations
Replication
Signal processing
Simulation
Stochastic models
Surrogate model
ISSN:0040-1706, 1537-2723
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:We investigate the merits of replication, and provide methods for optimal design (including replicates), with the goal of obtaining globally accurate emulation of noisy computer simulation experiments. We first show that replication can be beneficial from both design and computational perspectives, in the context of Gaussian process surrogate modeling. We then develop a lookahead-based sequential design scheme that can determine if a new run should be at an existing input location (i.e., replicate) or at a new one (explore). When paired with a newly developed heteroscedastic Gaussian process model, our dynamic design scheme facilitates learning of signal and noise relationships which can vary throughout the input space. We show that it does so efficiently, on both computational and statistical grounds. In addition to illustrative synthetic examples, we demonstrate performance on two challenging real-data simulation experiments, from inventory management and epidemiology. Supplementary materials for the article are available online.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0040-1706
1537-2723
DOI:10.1080/00401706.2018.1469433