Automatic Gradient Estimation for Calibrating Crowd Models with Discrete Decision Making

Andelfinger, Philipp and Kreikemeyer, Justin N. (2024) Automatic Gradient Estimation for Calibrating Crowd Models with Discrete Decision Making. In: 24th International Conference on Computational Science (ICCS 2024), 02-04 Jul 2024, Málaga, Spain. Proceedings, published by Springer Nature Switzerland, Cham, pp. 227-241.

Full text not available from this repository.
Official URL:


Recently proposed gradient estimators enable gradient descent over stochastic programs with discrete jumps in the response surface, which are not covered by automatic differentiation (AD) alone. Although these estimators’ capability to guide a swift local search has been shown for certain problems, their applicability to models relevant to real-world applications remains largely unexplored. As the gradients governing the choice in candidate solutions are calculated from sampled simulation trajectories, the optimization procedure bears similarities to metaheuristics such as particle swarm optimization, which puts the focus on the different methods’ calibration progress per function evaluation. Here, we consider the calibration of force-based crowd evacuation models based on the popular Social Force model augmented by discrete decision making. After studying the ability of an AD-based estimator for branching programs to capture the simulation’s rugged response surface, calibration problems are tackled using gradient descent and two metaheuristics. As our main insights, we find 1) that the estimation’s fidelity benefits from disregarding large jumps inherent to the Social Force model, and 2) that the common problem of inferring a parameter’s posterior distribution given some data obviates the need for AD across the Social Force calculations, allowing gradient descent to excel.

Item Type: Conference or Workshop Item (Paper)
Additional Information: Computational Science -- ICCS 2024, ed. by Franco,L. et al. LNCS, Vol.14836
Projects: SODA