Epidemiological Model Calibration via Graybox Bayesian Optimization
Puhua Niu, Byung-Jun Yoon, Xiaoning Qian · 2024
Read Epidemiological Model Calibration via Graybox Bayesian Optimization by Puhua Niu, Byung-Jun Yoon, Xiaoning Qian (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-07193.txt.
Introduction
Puhua Niu, Byung-Jun Yoon, and Xiaoning Qian wrote this 2024 paper on fitting epidemic models to data when each model run is expensive. Compartmental models, the workhorses that sort populations into susceptible, infected, and recovered, must be calibrated before they forecast, and standard Bayesian optimization treats the model as a pure black box. Their graybox scheme opens the box partway.
A Gaussian process stands in for the costly model, while the known functional structure of the compartment equations steers the search, sharpening calibration. A second, decoupled variant splits the decision steps to exploit the decomposability of the model further. On synthetic outbreaks and real COVID-19 data the graybox methods converged faster and landed lower error than plain Bayesian optimization. The authors offer the approach for the next expensive model family, epidemic or not.
Catalog record
| CREATOR | Puhua Niu, Byung-Jun Yoon, Xiaoning Qian |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 47 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2412-07193
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Epidemiological Model Calibration via Graybox Bayesian Optimization Puhua Niu, Byung-Jun Yoon, Xiaoning Qian Abstract In this study, we focus on developing efficient calibration methods via Bayesian decision-making for the family of compartmental epidemiological models. The existing calibration methods usually assume that the compartmental model is cheap in terms of its output and gradient evaluation, which may not hold in practice when extending them to more general settings. Therefore, we introduce model calibration methods based on a “graybox” Bayesian optimization (BO) scheme, more efficient calibration for general epidemiological models. This approach uses Gaussian processes as a surrogate to the expensive model, and leverages the functional structure of the compartmental model to enhance calibration performance. Additionally, we develop model calibration methods via a decoupled decision-making strategy for BO, which further exploits the decomposable nature of the functional structure. The calibration efficiencies of the multiple proposed schemes are evaluated based on various data generated by a compartmental model mimicking real-world epidemic processes, and real-world COVID-19 datasets. Experimental results demonstrate that our proposed graybox variants of BO schemes can efficiently calibrate computationally expensive models and further improve the calibration performance measured by the logarithm of mean square errors and achieve faster performance convergence in terms of BO iterations. We anticipate that the proposed calibration methods can be extended to enable fast calibration of more complex epidemiological models, such as the agent-based models. I Introduction Pandemics like the recent coronavirus disease (COVID-19) have shown enormous impacts on public health worldwide. The development of computational epidemiological models is crucial for gaining better quantitative understanding of the disease spread and enabling swift decision-making to design effective mitigation strategies. In fact, well-calibrated epidemiological models can serve as a useful guide for forecasting and quantifying the epidemic risk and implementing effective public health measures to control the spread of epidemic diseases [ 2 , 9 ] . Compartmental models, such as the the commonly adopted SIR (Susceptible-Infectious-Recovered) model, constitute an important family of population-based epidemiological models. Models in this family take the form of Ordinary Differential Equation (ODE) systems that capture the general trends and dynamics in the subpopulations represented by different compartments. In this work, we focus on this family of models while considering a new “Quarantined” state represented by an additional compartment. This new state aims at better mimicking the dynamic trajectories of the epidemic spreads, typical in airborne diseases such as seasonal flu and COVID-19, resulting in a four-compartment SIQR model [ 25 ] . To accurately model real-world physical processes, it is crucial for computer models to fine-tune their model parameters based on observed data to capture inherent attributes of the physical processes, which can not be directly or easily measured by means of physical experiments. For instance, material properties, such as porosity and permeability are important computer inputs in computational material simulations, which cannot be measured directly in physical experiments. In the applied mathematics and computational science literature, the methods used to identify those parameters are called model calibration techniques and the resulting parameters are called calibration parameters [ 22 , 28 ] . The basic idea of calibration is to find the combination of the calibration parameters, under which the simulated computer outputs align with the observed physical data. A well-calibrated SIQR model may allow people to make more accurate predictions of pandemic risk and enable them to make better-informed decisions on how to mitigate. The existing method often assumes that the computer model is cheap [ 33 ] . This means that: the explicit form of the optimization objective f(x) for calibration parameters x is known and we can evaluate the first- or second-order derivatives of f(x) with respect to (w.r.t.) x . For an ODE system, deriving the explicit form of f(x) and \nabla_{x}f(x) can be infeasible when the ODE is non-linear. Recent advances show that sensitivity analysis allows us to evaluate the gradient \nabla_{x}f(x) and perform the gradient-descent optimization method for the estimation of x [ 13 , 19 ] . Correspondingly, physics-informed machine learning [ 21 ] is drawing increasing attention as it integrates data and mathematical models into deep neural networks or other regression models by enforcing fundamental physical laws. In real-world scenarios when calibrating a computer simulation software, we are usually blind to the explicit forms of both f(x) and \nabla_{x}f(x) and we may be able to access only the output of f(x) . We classify this type of computer model as expensive . Besides, even if we know the explicit form of f(x) and \nabla_{x}f(x) , there are usually some key hyper-parameters x^{\prime} , which are fixed during gradient-based optimization. Given x , the explicit form of f_{x^{\prime}} is intractable in most cases. Therefore, f_{x^{\prime}} can also be classified as expensive . Hence, it is valuable to pay attention to the calibration problem for such expensive computer models. Bayesian Optimization (BO) is one conventional approach for such scenario when evaluating f(x) is expensive, where we use a surrogate model to approximate f(x) and the optimization process is sequential. In each BO round, we make a Bayesian decision so that some expected utility is maximized and then the corresponding output is queried from the computer model based on this decision to refine the surrogate model to guide the optimization of the objective function f(x) [ 18 , 15 ] . This paper concentrates specifically on the calibration of the SIQR model based on BO and is organized as follows. First, we explain the configurations of the SIQR model and the model calibration setup. Then we introduce our calibration methods based on a “graybox BO” approach, instead of the usual “blackbox BO”, where the experts’ prior knowledge about the computer models is integrated into the BO formulation to improve its optimization performance. Finally, we explain how the graybox BO works for decoupled decision-making, accompanied by a new acquisition function. In our experiments, we evaluate the performance of our proposed graybox BO-based model calibration methods and compare them with traditional blackbox BO-based implementations to demonstrate the benefits of integrating prior model knowledge into the calibration process. II Preliminaries Fig. 1 : Diagram depicting an SIQR compartmental epidemiological model. Each arrow indicates that the population rise of the ending compartment is caused by the population decrease of the starting compartment. Fig. 2 : Function network structure of the SIQR model calibration, where the two bold arrows indicate that x and g are the parent node and child node of y_{1:4} respectively. II-A SIQR model In this model, people in a given population are classified into four compartments: Susceptible (S) , Infectious (I) , Recovered (R) , and Quarantined (Q) . Those who have not been infected are considered susceptible. Transmission occurs between susceptible and infected individuals. The number of symptomatic patients, equivalent to the number of infected individuals, decreases through treatment and/or quarantine. Recovered individuals are assumed to remain immune to further infection. Therefore, all people will be in the recovered group after an adequate period. The dynamics of the SIQR model are given by the following set of ordinary differential equations (ODEs) [ 25 ] : \left\{\begin{matrix}\nabla_{t}S(t)=&-\beta(t)S(t)\\ \nabla_{t}I(t)=&\beta(t)S(t)-\lambda(t)I(t)-\gamma I(t)\\ \nabla_{t}R(t)=&\gamma I(t)+\delta Q(t)\\ \nabla_{t}Q(t)=&\lambda(t)I(t)-\delta Q(t)\\ \end{matrix}\right. (1) where S(t) , I(t) , Q(t) and R(t)\in[0,N] are the populations of compartments