Adaptation of uncertainty-penalized Bayesian information criterion for parametric partial differential equation discovery
This work addresses the challenge of parametric PDE discovery for researchers in physics and engineering, offering an incremental improvement by adapting an existing criterion to handle parametric dependencies more efficiently without expensive simulations.
The authors tackled the problem of identifying governing parametric partial differential equations (PDEs) from noisy data, where conventional criteria often select overly complex models. They introduced an extended uncertainty-penalized Bayesian information criterion (UBIC) that accurately identifies the true number of terms and their varying coefficients, even with noise, as demonstrated in numerical experiments on canonical PDEs.
Data-driven discovery of partial differential equations (PDEs) has emerged as a promising approach for deriving governing physics when domain knowledge about observed data is limited. Despite recent progress, the identification of governing equations and their parametric dependencies using conventional information criteria remains challenging in noisy situations, as the criteria tend to select overly complex PDEs. In this paper, we introduce an extension of the uncertainty-penalized Bayesian information criterion (UBIC), which is adapted to solve parametric PDE discovery problems efficiently without requiring computationally expensive PDE simulations. This extended UBIC uses quantified PDE uncertainty over different temporal or spatial points to prevent overfitting in model selection. The UBIC is computed with data transformation based on power spectral densities to discover the governing parametric PDE that truly captures qualitative features in frequency space with a few significant terms and their parametric dependencies (i.e., the varying PDE coefficients), evaluated with confidence intervals. Numerical experiments on canonical PDEs demonstrate that our extended UBIC can identify the true number of terms and their varying coefficients accurately, even in the presence of noise. The code is available at \url{https://github.com/Pongpisit-Thanasutives/parametric-discovery}.