Agent-based epidemic models (ABMs) encode behavioral and policy heterogeneity but are too slow for nightly hospital planning. We develop county-ready surrogates that learn directly from exascale ABM trajectories using Universal Differential Equations (UDEs): mechanistic SEIR-family ODEs with a neural-parameterized contact rate κφ(u,t) (no additive residual). Our contributions are threefold: we adapt multiple shooting and an observer-based prediction-error method (PEM) to stabilize identification of neural-augmented epidemiological dynamics across intervention-driven regime shifts; we enforce positivity and mass conservation and show the learned contact-rate parameterization yields a well-posed vector field; and we quantify accuracy, calibration, and compute against ABM ensembles and UDE baselines. On a representative ExaEpi scenario, PEM-UDE reduces mean MSE by 77% relative to single-shooting UDE (3.00 vs. 13.14) and by 20% relative to MS-UDE (3.75). Reliability improves in parallel: empirical coverage of ABM 10-90% and 25-75% bands rises from 0.68/0.43 (UDE) and 0.79/0.55 (MS-UDE) to 0.86/0.61 with PEM-UDE and 0.94/0.69 with MS+PEM-UDE, indicating calibrated uncertainty rather than overconfident fits. Inference runs in seconds on commodity CPUs (20-35 s per ∼90-day forecast), enabling nightly ''what-if'' sweeps on a laptop. Relative to a ∼100 CPU-hour ABM reference run, this yields ∼104× lower wall-clock per scenario. This closes the realism-cadence gap, supports threshold-aware decision-making (e.g., maintaining ICU occupancy <75%), preserves mechanistic interpretability, and enables calibrated, risk-aware scenario planning on standard institutional hardware. Beyond epidemics, the ABM→UDE recipe provides a portable path to distill agent-based simulators into fast, trustworthy surrogates for other scientific domains.
Figures & tables
Figure 1: Pipeline overview (one-scenario evaluation). Data: ExaEpi (ABM) generates a single representative SEInsIsIaDR outbreak trajectory (Dataset 1). Known constraints: population balance, compartmental flow, and mechanistic ODE structure constrain learning. Learn dynamics: a neural-parameterized UDE is trained with stabilization strategies: Vanilla UDE, MS-UDE, PEM-UDE, and MS+PEM-UDE. Ensemble robustness: each method is trained as a 100-seed ensemble to quantify sensitivity to initialization and optimization stochasticity. Inference: the trained UDE supports fast “what-if” forward simulation by varying x0 and contact schedules κ(t) on commodity hardware.
Figure 2: Epochs v MSE for all methods (100-seed avg)
Method
Final MSE (Dataset 1)
UDE (baseline)
16.85
MS-UDE
3.76
PEM-UDE
3.01
MS+PEM-UDE
2.93
Table 1: Final mean-squared error (MSE) on Dataset 1 (average over 100 seeds). Lower is better.
Method
Latency
UDE (baseline)
21 s ( 24 s )
MS-UDE (inference)
23 s ( 26 s )
PEM-UDE ( K=0 )
31 s ( 35 s )
MS+PEM-UDE ( K=0 )
32 s ( 35 s )
ExaEpi ABM (reference)
100 h = 360,000 s
Table 2: Per-trajectory inference latency on Apple M4 (single thread, warmed session, daily saveat , moderate tolerances) compared with an ExaEpi ABM reference run. Latency values are medians with p90 in parentheses (seconds for UDEs; hours for ABM). The ABM reference corresponds to approximately 100 CPU-hours of execution.
(a) Exposed ( E ).
(b) Symptomatic ( Is ).
(c) Asymptomatic ( Ia ).
(d) Presymptomatic ( I∼s ).
(e) Deaths ( D ).
(f) Never-infected / susceptible-like pool.
Method
10–90 % Coverage
25–75 % Coverage
UDE (baseline)
0.68
0.43
MS-UDE
0.79
0.55
PEM-UDE
0.86
0.61
MS+PEM-UDE (ensemble)
0.94
0.69
Table 3: Mean empirical coverage of the 10–90 % and 25–75 % ABM quantile bands on Dataset 1. Higher values indicate better calibration to the ensemble variability.
Compartment
MS+PEM vs. PEM
MS+PEM vs. MS
MS+PEM vs. UDE
Exposed ( E )
−2.3%
−8.6%
−86.3%
Presymptomatic ( I∼s )
−3.4%
−7.6%
−82.6%
Symptomatic ( Is )
−5.6%
−10.4%
−91.2%
Asymptomatic ( Ia )
−6.8%
−9.1%
−87.4%
Never-infected / Susceptible-like pool ( S )
−1.5%
−5.4%
−70.1%
Deaths ( D )
−4.3%
−7.3%
−68.4%
Table 4: Relative change in per-compartment MSE of MS+PEM-UDE compared to PEM-UDE, MS-UDE, and Vanilla UDE on Dataset 1. Negative values indicate MS+PEM-UDE achieves lower MSE.
This paper presents a hybrid modeling approach that couples an Agent-Based Model (ABM) with a partial differential equation (PDE) model in an epidemic setting to simulate the spatial spread of infectious diseases using a compartmental structure with seven health states. The goal is to reduce the computational complexity of a full-ABM by introducing a coupled ABM-PDE model that offers significantly faster simulations while maintaining comparable accuracy. Our results demonstrate that the hybrid model not only reduces the overall simulation runtime (defined as the number of runs required for stable results multiplied by the duration of a single run) but also achieves smaller errors across both 25% and 100% population samples. The coupling mechanism ensures consistency at the model interface: agents crossing from the ABM into the PDE domain are removed and represented as density contributions, while surplus density in the PDE domain is used to generate agents with plausible trajectories derived from mobile phone data. We evaluate the hybrid model using real-world mobility and infection data for the Berlin-Brandenburg region in Germany, showing that it captures the core epidemiological dynamics while enabling efficient large-scale simulations. These results demonstrate that the proposed ABM-PDE framework provides a robust and computationally efficient alternative to full-scale agent-based simulations, making it suitable for realistic epidemic modeling and scenario analysis.
Systematic exploration of Agent-Based Models (ABMs) is challenged by the curse of dimensionality and their inherent stochasticity. We present a multi-stage pipeline integrating the systematic design of experiments with machine learning surrogates. Using a predator-prey case study, our methodology proceeds in two steps. First, an automated model-based screening identifies dominant variables, assesses outcome variability, and segments the parameter space. Second, we train Machine Learning models to map the remaining nonlinear interaction effects. This approach automates the discovery of unstable regions where system outcomes are highly dependent on nonlinear interactions between many variables. Thus, this work provides modelers with a rigorous, hands-off framework for sensitivity analysis and policy testing, even when dealing with high-dimensional stochastic simulators.
Paul Saves, Matthieu Mastio, Nicolas Verstaevel +1
IRIT, Université Toulouse Capitole, Toulouse, France
Agent-based modeling (ABM) has the capability to model millions of individuals and their interactions, which is useful for policy making. However, ABMs have traditionally relied on static prior, which prevents the models from adapting to real-time changes. Our research provides a novel approach to addressing this information gap. Large language models (LLMs) offer new opportunities to predict human decision-making. Here, we introduce a scalable Hybrid Agent-based and Language-driven Epidemic (HALE) modeling framework that leverages LLMs to predict human decision-making in an ABM simulation. As a proof-of-concept, we use HALE to simulate COVID-19 and its effects in Salt Lake County, UT.
Sifat Afroj Moon, Dakotah Maguire, Adam Spannaus +5
Oak Ridge National Laboratory · Computational Science and Engineering Division, Oak Ridge National Laboratory, Oak Ridge, TN 37830, USA · Geospatial Science and Human Security Division, Oak Ridge National Laboratory, Oak Ridge, TN 37830, USA +1