Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 473 - Design of Experiments and Advanced Analytics
Type: Contributed
Date/Time: Thursday, August 6, 2020 : 10:00 AM to 2:00 PM
Sponsor: Quality and Productivity Section
Abstract #313847
Title: Batch-Sequential Design and Heteroskedastic Surrogate Modeling for Delta Smelt Conservation
Author(s): Boya Zhang* and Robert Gramacy and Eric Smith and Leah R Johnson and Kenny Rose
Companies: Virginia Tech and Virginia Tech and Virginia Tech and Virginia Polytechnic and State University and University of Maryland
Keywords: Computer experiment; Gaussian process; heteroskedastic noise; Surrogate model; Replication
Abstract:

Delta smelt are a small endangered fish whose fate is intimately linked with water management practice in the Sacramento river delta system, and who more broadly serve as a barometer for environmental health in the San Francisco Bay. Researchers have developed a stochastic, agent-based simulator to virtualize the system, with the goal of assisting in a study of delta smelt life cycles and to understand sensitivities to myriad natural variables and human interventions. However, the input configuration space is high-dimensional, running the simulator is time-consuming, and its noisy outputs change nonlinearly in both mean and variance. Getting enough runs to effectively learn input-output dynamics requires both a nimble modeling strategy and a parallel supercomputer evaluation. We propose a batch sequential design scheme, generalizing one-at-a-time variance-based active learning for hetGP surrogates, as a means of keeping multi-core cluster nodes fully engaged with expensive runs. Our acquisition strategy is carefully engineered to favor selection of replicates which boost statistical and computational efficiencies when training surrogates to isolate signal in high noise regions.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2020 program