Batch Sequential Experimental Design for Calibration of Stochastic Simulation Models

TL;DR

This paper introduces a batch sequential experimental design method for calibrating stochastic simulation models, improving efficiency by intelligently selecting simulation points to reduce uncertainty in posterior predictions.

Contribution

It proposes novel criteria for batch selection in sequential design, optimizing the calibration process of stochastic models in parallel computing environments.

Findings

Improved posterior prediction accuracy in simulated models.

Enhanced calibration efficiency with batch evaluation strategy.

Effective handling of noisy outputs in stochastic simulations.

Abstract

Calibration of expensive simulation models involves an emulator based on simulation outputs generated across various parameter settings to replace the actual model. Noisy outputs of stochastic simulation models require many simulation evaluations to understand the complex input-output relationship effectively. Sequential design with an intelligent data collection strategy can improve the efficiency of the calibration process. The growth of parallel computing environments can further enhance calibration efficiency by enabling simultaneous evaluation of the simulation model at a batch of parameters within a sequential design. This article proposes novel criteria that determine if a new batch of simulation evaluations should be assigned to existing parameter locations or unexplored ones to minimize the uncertainty of posterior prediction. Analysis of several simulated models and real-data experiments from epidemiology demonstrates that the proposed approach results in improved posterior predictions.