The Particle Swarm Optimization algorithm running on the Rastrigin function
I have done a lot of work, especially with Dr. Jeon-Young Kang, using spatially-explicit agent-based models for investigating disease spread. This work has examined using CyberGISX for reproducible and transparent modeling and calibration of ABMs.
Associated Publications
Journal Articles
2022
Particle Swarm Optimization for Calibration in Spatially Explicit Agent-Based Modeling
Journal of Artificial Societies and Social Simulation,
2022
A challenge in computational modeling of Agent-Based Models (ABMs) is the amount of time and resources required to tune a set of parameters for reproducing the observed patterns of phenomena being modeled. Well-tuned parameters are necessary for models to reproduce real-world multi-scale space-time patterns, but calibration is often computationally intensive and time consuming. Particle Swarm Optimization (PSO) is a swarm intelligence optimization algorithm that has found wide use for complex optimization including nonconvex and noisy problems. In this study, we propose to use PSO for calibrating parameters in ABMs. We use a spatially explicit ABM of influenza transmission based in Miami, Florida, USA as a case study. Furthermore, we demonstrate that a standard implementation of PSO can be used out-of-the-box to successfully calibrate models and out-performs Monte Carlo in terms of optimization and efficiency.
2021
An Integrated Framework of Global Sensitivity Analysis and Calibration for Spatially Explicit Agent-Based Models
Transactions in GIS,
2021
Abstract Calibration of agent-based models (ABMs) is a major challenge due to the complex nature of the systems being modeled, the heterogeneous nature of geographical regions, the varying effects of model inputs on the outputs, and computational intensity. Nevertheless, ABMs need to be carefully tuned to achieve the desirable goal of simulating spatiotemporal phenomena of interest, and a well-calibrated model is expected to achieve an improved understanding of the phenomena. To address some of the above challenges, this article proposes an integrated framework of global sensitivity analysis (GSA) and calibration, called GSA-CAL. Specifically, variance-based GSA is applied to identify input parameters with less influence on differences between simulated outputs and observations. By dropping these less influential input parameters in the calibration process, this research reduces the computational intensity of calibration. Since GSA requires many simulation runs, due to ABMs’ stochasticity, we leverage the high-performance computing power provided by the advanced cyberinfrastructure. A spatially explicit ABM of influenza transmission is used as the case study to demonstrate the utility of the framework. Leveraging GSA, we were able to exclude less influential parameters in the model calibration process and demonstrate the importance of revising local settings for an epidemic pattern in an outbreak.
Conference Proceedings
2019
CyberGIS-Jupyter for Spatially Explicit Agent-based Modeling: A Case Study on Influenza Transmission
GeoSim ’19: Proceedings of the 2nd ACM SIGSPATIAL International Workshop on GeoSpatial Simulation,
2019
Despite extensive efforts on achieving reproducible agent-based models (ABMs) to improve the capability of this widely adopted methodology, it remains challenging to reproduce and replicate pre-existing ABMs, due to a number of factors such as diverse computing resources and ABMs platforms. In this study, we propose to employ CyberGIS-Jupyter for spatially explicit ABMs. CyberGIS-Jupyter is a cyberGIS framework to achieve data-intensive, reproducible, and scalable geospatial analytics using Jupyter Notebook based on advanced cyberinfrastructure. Influenza transmission in the city of Miami, Florida, USA was used as a case study. In the model, Influenza is transmitted through the contact networks of individual human agents, which are constructed based on commuting behaviors. CyberGIS-Jupyter can support one not only to conduct collaborative and transparent modeling, but also to perform modeling simulation on advanced cyberinfrastructure resources. It may contribute to boosting the reproducibility and replicability of ABMs.
