Trajectory-Oriented Optimization of Stochastic Epidemiological Models

Arindam Fadikar, Nicholson Collier, Abby Stevens, Jonathan Ozik, Mickael Binois, and Kok Ben Toh

2023 Winter Simulation Conference (WSC), pp. 1244-1255, 2023

IEEE Xplore

Trajectory-oriented optimization for stochastic epidemiological models

Abstract

This paper develops trajectory-oriented optimization methods for stochastic epidemic models, focusing on matching full temporal dynamics rather than single summary targets.

The approach frames calibration as a trajectory-matching problem and uses optimization strategies tailored to noisy, computationally expensive simulation outputs.

By prioritizing agreement with time-varying epidemic patterns, the method improves calibration fidelity and supports more realistic scenario analysis for decision support.

Citation

BibTeX

@inproceedings{fadikar2023too,
  title={Trajectory-Oriented Optimization of Stochastic Epidemiological Models},
  author={Fadikar, Arindam and Collier, Nicholson and Stevens, Abby and Ozik, Jonathan and Binois, Mickael and Toh, Kok Ben},
  booktitle={2023 Winter Simulation Conference (WSC)},
  pages={1244--1255},
  year={2023},
  organization={IEEE},
  doi={10.1109/WSC60868.2023.10408258}
}