Figure 2
From: Algorithmic discovery of dynamic models from infectious disease data
Comparison of the simulated SIR model with additive noise of \(\varepsilon =1\times {10}^{-7}\) (top) and \(\varepsilon =2\times {10}^{-5}\) (bottom) with the corresponding coefficients of terms in the discovered model. \({\beta }_{0}=8\)/wk, \({\beta }_{1}=0.25\), \(\gamma =0.1\)/wk, \(\mu =\nu =5.4795\times {10}^{-5}\)/wk. The results in the table display the SINDy-discovered coefficients of the corresponding terms in (Eqs. 1–3). In panel (a) the Data and SINDy curves are very closely overlapping, hence the appearance of only a single blue curve.
