arXiv subjects

David O'Gara

Publications and source records attributed to David O'Gara.

5 recordsLinked to original sources

arXiv2026

Trajectory-Oriented Optimization Via Adaptive Thompson Sampling And Grid Refinement: A Tutorial With The ADAPTIVE\_TS Package

Stochastic simulators are increasingly used to expand the frontier of scientific knowledge and inform decision-making across real-world contexts. Simulator calibration, a process by which internal model inputs are tuned to match some external criteria, usually in the form of observed data, is a key step in model design and validation. Epidemiological simulators present an especially compelling use case, as evidenced by the recent COVID-19 pandemic. Among several calibration paradigms, trajectory-oriented optimization is an emerging approach that does not require assumptions on the stochastic behavior of the simulator replicates and is particularly effective at identifying trajectories through the lens of errors between the simulator and observed data, especially when combined with Bayesian optimization. We present a tutorial on trajectory-oriented optimization with \texttt{adaptive\_ts}, an open-source Python package. We also provide a series of worked examples on an accompanying webpage.

stat.CO↗

arXiv2026

Staying on Track: Efficient Trajectory Discovery with Adaptive Batch Sampling

Bayesian optimization (BO) is a powerful framework for estimating parameters of expensive simulation models, particularly in settings where the likelihood is intractable and evaluations are costly. In stochastic models every simulation is run with a specific parameter set and an implicit or explicit random seed, where each parameter set and random seed combination generates an individual realization, or trajectory, sampled from an underlying random process. Existing BO approaches typically rely on summary statistics over the realizations, such as means, medians, or quantiles, potentially limiting their effectiveness when trajectory-level information is desired. We propose a trajectory-oriented BO method that incorporates a Gaussian process surrogate using both input parameters and random seeds as inputs, enabling direct inference at the trajectory level. Using a common random number approach, we define a surrogate-based likelihood over trajectories and introduce an adaptive Thompson Sampling algorithm that refines a fixed-size input grid through likelihood-based filtering and Metropolis-Hastings-based densification. This approach concentrates computation on statistically promising regions of the input space while balancing exploration and exploitation. We apply the method to stochastic epidemic models, a simple compartmental and a more computationally demanding agent-based model, demonstrating improved sampling efficiency and faster identification of data-consistent trajectories relative to parameter-only inference.

stat.ME↗

arXiv2025

Start from the End: A Framework for Computational Policy Exploration to Inform Effective and Geospatially Consistent Interventions applied to COVID-19 in St. Louis

Mathematical models are a powerful tool to study infectious disease dynamics and intervention strategies against them in social systems. However, due to their detailed implementation and steep computational requirements, practitioners and stakeholders are typically only able to explore a small subset of all possible intervention scenarios, a severe limitation when preparing for disease outbreaks. In this work, we propose a parameter exploration framework utilizing emulator models to make uncertainty-aware predictions of high-dimensional parameter spaces and identify large numbers of feasible response strategies. We apply our framework to a case study of a large-scale agent-based disease model of the COVID-19 ``Omicron wave'' in St. Louis, Missouri that took place from December 2021 to February 2022. We identify large numbers of response strategies that would have been estimated to have reduced disease spread by a substantial amount. We also identify policy interventions that would have been able to reduce the geospatial variation in disease spread, which has additional implications for designing thoughtful response strategies.

stat.AP↗

arXiv2024

Improving Policy-Oriented Agent-Based Modeling with History Matching: A Case Study

Advances in computing power and data availability have led to growing sophistication in mechanistic mathematical models of social dynamics. Increasingly these models are used to inform real-world policy decision-making, often with significant time sensitivity. One such modeling approach is agent-based modeling, which offers particular strengths for capturing spatial and behavioral realism, and for in-silico experiments (varying input parameters and assumptions to explore their downstream impact on key outcomes). To be useful in the real world, these models must be able to qualitatively or quantitatively capture observed empirical phenomena, forming the starting point for subsequent experimentation. Computational constraints often form a significant hurdle to timely calibration and policy analysis in high resolution agent-based models. In this paper, we present a technical solution to address this bottleneck, substantially increasing efficiency and thus widening the range of utility for policy models. We illustrate our approach with a case study using a previously published and widely used epidemiological model.

stat.AP↗

arXiv2023

TRACE-Omicron: Policy Counterfactuals to Inform Mitigation of COVID-19 Spread in the United States

The Omicron wave was the largest wave of COVID-19 pandemic to date, more than doubling any other in terms of cases and hospitalizations in the United States. In this paper, we present a large-scale agent-based model of policy interventions that could have been implemented to mitigate the Omicron wave. Our model takes into account the behaviors of individuals and their interactions with one another within a nationally representative population, as well as the efficacy of various interventions such as social distancing, mask wearing, testing, tracing, and vaccination. We use the model to simulate the impact of different policy scenarios and evaluate their potential effectiveness in controlling the spread of the virus. Our results suggest the Omicron wave could have been substantially curtailed via a combination of interventions comparable in effectiveness to extreme and unpopular singular measures such as widespread closure of schools and workplaces, and highlight the importance of early and decisive action.

cs.SI↗