Equation Discovery with Bayesian Spike-and-Slab Priors and Efficient Kernels

arXiv:2310.05387v2 Announce Type: replace
Abstract: Discovering governing equations from data is important to many scientific and engineering applications. Despite promising successes, existing methods are still challenged by data sparsity and noise issues, both of which are ubiquitous in practice. Moreover, state-of-the-art methods lack uncertainty quantification and/or are costly in training. To overcome these limitations, we propose a novel equation discovery method based on Kernel learning and BAyesian Spike-and-Slab priors (KBASS). We use kernel regression to estimate the target function, which …

abstract applications art arxiv bayesian cs.lg data discovery engineering equation noise practice quantification scientific sparsity state stat.ml training type uncertainty