From Prediction to Prescription: Evolutionary Optimization of Nonpharmaceutical Interventions in the COVID-19 Pandemic

Abstract

Several models have been developed to predict how the COVID-19 pandemic spreads, and how it could be contained with nonpharmaceutical interventions, such as social distancing restrictions and school and business closures. This article demonstrates how evolutionary AI can be used to facilitate the next step, i.e., determining most effective intervention strategies automatically. Through evolutionary surrogate-assisted prescription, it is possible to generate a large number of candidate strategies and evaluate them with predictive models. In principle, strategies can be customized for different countries and locales, and balance the need to contain the pandemic and the need to minimize their economic impact. Early experiments suggest that workplace and school restrictions are the most important and need to be designed carefully. They also demonstrate that results of lifting restrictions can be unreliable, and suggest creative ways in which restrictions can be implemented softly, e.g., by alternating them over time. As more data becomes available, the approach can be increasingly useful in dealing with COVID-19 as well as possible future pandemics.

EpidemiOptim: A Toolbox for the Optimization of Control Policies in Epidemiological Models

“…This latter paper also proposes a stochastic agent-based model called VIPER (Virus-Individual-Policy-EnviRonment) allowing to compare the optimization results on variations of the demographics and geographical distribution of population. Finally, Miikkulainen et al (2020) proposed an original approach using Evolutionary Surrogate-assisted Prescription (ESP). In this approach, a recurrent neural network (the Predictor) was trained with publicly available data on infections and NPIs in several countries and applied to predicting how the pandemic will unfold in them in the future.…”

Deciphering the COVID-19 Health Economic Dilemma (HED): A Scoping Review

“…Currently, as apparent from reviews such as in Musulin et al (2021) [81] and Tseng et al (2020) [82], there is a vast volume of studies on virus spread prediction and control models; however, there is a lack of studies that combine such models with economic models to reveal the HED related tradeoffs. Some initial research attempts towards this goal can be found in recent works such as in Yousefpour et al (2020) [83], Salgotra et al (2021) [84] and Miikkulainen et al (2021) [85]. It is expected that the current review will not only support policy makers but will also provide researchers on the development of related decision-support-systems with comprehensive information on the various aspects of the HED.…”

Open Data Science to Fight COVID-19: Winning the 500k XPRIZE Pandemic Response Challenge

“…Several works evaluated the effectiveness of NPIs: see [21,20] for studies in Italy, Taiwan and Malaysia or [10,6] for recent studies in Europe. Finally, Miikkulainen et al propose a neuroevolution approach to identify a Pareto-optimal set of NPIs [17], that was recommended during the Challenge.…”

“…The baseline or standard predictor was provided by the Challenge organizers [17]. It consists of two parallel LSTMs, one to model the context -given by the R j n -and the other to model the actions (A j n ) applied on day n in GEO j.…”

“…, A j n−1 ). In our experiments we set T = 21, similarly to [17]. Such time window mitigates the noise due to how different GEOs report cases (e.g.…”