cs.MAApr 2, 2026

Optimizing Interventions for Agent-Based Infectious Disease Simulations

Authors: Anja Wolpers, Johannes Ponge, Adelinde M. Uhrmacher

Organizations: University of Rostock, Rostock, Germany · University of Münster, Münster, Germany*

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Non-pharmaceutical interventions (NPIs) are commonly used tools for controlling infectious disease transmission when pharmaceutical options are unavailable. Yet, identifying effective interventions that minimize societal disruption remains challenging. Agent-based simulation is a popular tool for analyzing the impact of possible interventions in epidemiology. However, automatically optimizing NPIs using agent-based simulations poses a complex problem because, in agent-based epidemiological models, interventions can target individuals based on multiple attributes, affect hierarchical group structures (e.g., schools, workplaces, and families), and be combined arbitrarily, resulting in a very large or even infinite search space. We aim to support decision-makers with our Agent-based Infectious Disease Intervention Optimization System (ADIOS) that optimizes NPIs for infectious disease simulations using Grammar-Guided Genetic Programming (GGGP). The core of ADIOS is a domain-specific language for expressing NPIs in agent-based simulations that structures the intervention search space through a context-free grammar. To make optimization more efficient, the search space can be further reduced by defining constraints that prevent the generation of semantically invalid intervention patterns. Using this constrained language and an interface that enables coupling with agent-based simulations, ADIOS adopts the GGGP approach for simulation-based optimization. Using the German Epidemic Micro-Simulation System (GEMS) as a case study, we demonstrate the potential of our approach to generate optimal interventions for realistic epidemiological models

Figures & tables

Figure 2: Example intervention tree for the intervention ”Symptomatic students have to self-isolate, and their school has to close”.
Figure 3: The parse tree for the example intervention tree in Figure 2 . In the brackets are the rules that are applied to the token above to reach the token below.
Figure 4: Filtering a set repeatedly with the same filter will repeatedly result in the same set (here, ASA_{S} ). Banning repeating identical filters prevents these semantically pointless interventions.
Figure 5: Example of how the bans impact the expansion of symbols: Before the rule 5 ∈Q\in Q for {\color[rgb]{0,0.5,0.5}p_{a}?} is applied, it is filtered by the bans attached to the previous intervention tree edge. An intervention tree edge corresponds to a parse tree’s sub-tree starting with a node labeled with starta and ending before the next node labeled starta. Here, the nodes corresponding to the two intervention tree edges are grouped in the gray bubbles.
Figure 6: The steps in our GGGP algorithm. First, a new generation of NPIs is generated based on our language’s grammar. Then, the NPIs are evaluated in the context of a given goal function. Next, based on the NPIs’ fitness values, they are evolved into a new generation. The green intervention has a low fitness and is therefore not part of the new generation. The blue one has the best fitness and is cloned into the next generation, then crossed with the yellow one. In addition, one intervention in the new generation has been mutated, which is visualized by the pink node that has not yet been part of the population.
Figure 7: Example for a crossover between two intervention parse trees. With the given ban, the highlighted starta node in the left tree and the starta node in the right tree with the dashed outline are not a viable option for crossover. However, the combination of the left starta node and the other highlighted right starta node with the continuous outline is a legal crossover pair.
Figure 8: Observed Parameter values for the fittest NPI per scenario and for a baseline simulation without any NPIs. Scenario 1: Minimize sick days; Scenario 2: Minimize lost school days per student; Scenario 3: Minimize lost work days per worker; Scenario 4: combined goal function (lost_work_days + lost_school_days + 2 * sick_days).
Figure 9: The mean and max fitness of the NPIs over the generations of our genetic programming algorithm per scenario (with different goal functions). We define the fitness to be 1/(<target function>+1). Thus, the fitness in scenario four, where the goal function adds up the lost workdays, lost school days, and sick days, is generally lower than in the other scenarios, where only one metric makes up the goal function. Scenario 1: Minimize sick days; Scenario 2: Minimize lost school days per student; Scenario 3: Minimize lost work days per worker; Scenario 4: combined goal function (lost_work_days + lost_school_days + 2 * sick_days).
More in this topic

Agent-Based Modeling

Agent-based modeling simulates how interacting, heterogeneous entities and their environments produce patterns in social, economic, biological, and engineered systems.

See all 153 papers

Explore similar work

CardsList
Mar 19, 2026

Optimizing Resource-Constrained Non-Pharmaceutical Interventions for Multi-Cluster Outbreak Control Using Hierarchical Reinforcement Learning

Non-pharmaceutical interventions (NPIs), such as diagnostic testing and quarantine, are crucial for controlling infectious disease outbreaks but are often constrained by limited resources, particularly in early outbreak stages. In real-world public health settings, resources must be allocated across multiple outbreak clusters that emerge asynchronously, vary in size and risk, and compete for a shared resource budget. Here, a cluster corresponds to a group of close contacts generated by a single infected index case. Thus, decisions must be made under uncertainty and heterogeneous demands, while respecting operational constraints. We formulate this problem as a constrained restless multi-armed bandit and propose a hierarchical reinforcement learning framework. A global controller learns a continuous action cost multiplier that adjusts global resource demand, while a generalized local policy estimates the marginal value of allocating resources to individuals within each cluster. We evaluate the proposed framework in a realistic agent-based simulator of SARS-CoV-2 with dynamically arriving clusters. Across a wide range of system scales and testing budgets, our method consistently outperforms RMAB-inspired and heuristic baselines, improving outbreak control effectiveness by 20%-30%. Experiments on up to 40 concurrently active clusters further demonstrate that the hierarchical framework is highly scalable and enables faster decision-making than the RMAB-inspired method.

Xueqiao Peng, Andrew Perrault

The Ohio State University

HealthcareConstrained RLHierarchical RL
AskPDF
Sep 23, 2026

Driving Epidemic Models with AI Agents: the Epydemix Agent Framework

Artificial Intelligence agents based on large language models provide convenient natural language interfaces to scientific software, but reliability is not automatic. Here we introduce the Epydemix Agent Framework, an additive layer over Epydemix, an open-source Python library for stochastic compartmental epidemic modeling. The framework extends the library with four capabilities to facilitate interaction with an AI agent: discovery of available models and parameters, preventive validation of a declarative scenario specification, execution through tested library code, and inspectability of results. These capabilities let an agent handle the entire modeling process, from the natural-language description of the scenario to quantitative results, figures, and interpretation of findings without writing custom code. Each step reads input files and saves results in a separate output bundle, making the process auditable and reproducible. First, we show the end-to-end workflow with a case study comparing vaccination strategies for a novel respiratory virus. Second, we assessed the framework across 50 agent sessions and five modeling tasks by comparing the agent use of the framework against the direct use of the Python interface. The framework reduced turns, output tokens, and cost on most tasks, unless it trades resources for per-point reproducibility.

Nicolò Gozzi, Ciro Cattuto, Alessandro Vespignani

ISI Foundation, Turin, Italy · Network Science Institute, Northeastern University, Boston, MA, USA

AI Agents for Scientific DiscoveryLLM Agent HarnessesLLM Agent ReliabilitySpatial Epidemic Modeling
AskPDF
Oct 2, 2026

EpiWorld: Grounding LLM Policy Agents in Epidemiological World Models

Epidemic intervention policies are textual artefacts that human decision-makers interpret, justify, and revise through natural language, making large language models a natural candidate for epidemic policy reasoning. A naive LLM, however, lacks the epidemic dynamics needed to project intervention consequences, the quantitative surveillance signals required to assess severity, and the institutional constraints that define admissible actions. We present EpiWorld, a closed-loop framework that grounds an LLM policy actor in a learned action-conditioned epidemiological world model and a tiered skill library of public-health protocols, surveillance tools, and adaptive lessons accumulated through after-action analysis. Given a candidate intervention, the world model predicts regional epidemic evolution and enables fast counterfactual rollouts that provide feedback for policy selection and refinement. Outcomes of simulated futures are distilled into reusable lessons while protocol constraints remain fixed, allowing the decision process to improve without sacrificing interpretability or controllability. We evaluate both the world model and the end-to-end framework on retrospective COVID-19 and Influenza datasets: the world model achieves the best out-of-distribution Peak-MAE among all forecasting baselines, and the closed-loop framework reduces cumulative hospitalisation by up to 59% across datasets and by an average of ~16% across six LLM backbones, outperforming reinforcement-learning and optimal-control policy baselines.

Zeeshan Memon, Yiqi Su, Kai Shu +2

Department of Computer Science, Emory University, Atlanta, GA, USA · Department of Computer Science, Virginia Tech, Alexandria, VA, USA

Action-Conditioned World ModelsEpidemic Forecasting
AskPDF