From Model-Based Screening to Data-Driven Surrogates: A Multi-Stage Workflow for Exploring Stochastic Agent-Based Models

1. 1Identify dominant variables using automated model-based screening
2. 2Assess outcome variability and segment the parameter space
3. 3Develop data-driven surrogates using machine learning
4. 4Integrate surrogates into the workflow for efficient exploration

Data scientists and AI engineers on a team can benefit from this workflow to efficiently explore complex Agent-Based Models, while product managers can apply the insights gained to inform decision-making

💡 Integrating model-based screening with data-driven surrogates can overcome the challenges of exploring complex Agent-Based Models

🤖 Explore stochastic Agent-Based Models efficiently with a multi-stage workflow! 📈