Trajectory-Oriented Optimization Via Adaptive Thompson Sampling And Grid Refinement: A Tutorial With The ADAPTIVE\_TS Package

2026-05-22

AI Summary

This study addresses the challenge of calibrating stochastic simulators—such as epidemiological models—where matching observed data is difficult due to inherent randomness and the absence of prior assumptions about their stochastic behavior. To overcome this, the authors propose a trajectory-oriented calibration method that integrates adaptive Thompson sampling with a grid refinement strategy, efficiently optimizing model parameters by minimizing the discrepancy between simulated trajectories and observational data. The key innovation lies in its ability to operate without assuming any specific distributional form for the simulator’s repeated stochastic outputs, seamlessly combining Bayesian optimization with trajectory-level error evaluation. The approach is implemented in an open-source Python package, ADAPTIVE_TS. Experimental results across multiple epidemiological simulation scenarios demonstrate that the method successfully identifies parameter configurations yielding trajectories highly consistent with real-world observations, substantially improving both calibration accuracy and computational efficiency.

Abstract

Stochastic simulators are increasingly used to expand the frontier of scientific knowledge and inform decision-making across real-world contexts. Simulator calibration, a process by which internal model inputs are tuned to match some external criteria, usually in the form of observed data, is a key step in model design and validation. Epidemiological simulators present an especially compelling use case, as evidenced by the recent COVID-19 pandemic. Among several calibration paradigms, trajectory-oriented optimization is an emerging approach that does not require assumptions on the stochastic behavior of the simulator replicates and is particularly effective at identifying trajectories through the lens of errors between the simulator and observed data, especially when combined with Bayesian optimization. We present a tutorial on trajectory-oriented optimization with \texttt{adaptive\_ts}, an open-source Python package. We also provide a series of worked examples on an accompanying webpage.

Problem

• simulator calibration
• trajectory-oriented optimization
• stochastic simulators
• epidemiological simulators
• observed data

Innovation

• trajectory-oriented optimization
• adaptive Thompson sampling
• grid refinement
• stochastic simulator calibration
• Bayesian optimization