Physics-Informed Neural Networks and Functional Interpolation for Data-Driven Parameters Discovery of Epidemiological Compartmental Models

In this work, we apply a novel and accurate Physics-Informed Neural Network Theory of Functional Connections (PINN-TFC) based framework, called Extreme (X-TFC), for data-physics-driven parameters’ discovery problems modeled via Ordinary Differential Equations (ODEs). The proposed method merges the standard PINNs with functional interpolation technique named (TFC). particular, work focuses on capability X-TFC in solving inverse to estimate parameters governing epidemiological compartmental models deterministic approach. treated are Susceptible-Infectious-Recovered (SIR), Susceptible-Exposed-Infectious-Recovered (SEIR), Susceptible-Exposed-Infectious-Recovered-Susceptible (SEIRS). results show low computational times, high accuracy, effectiveness performing data-driven systems parametric ODEs using unperturbed perturbed data.