Marcus M. Noack

1 Introduction Gaussian processes (GPs) are a preeminent framework for stochastic function approximation, statistical modeling of real-world measurements, non-parametric and nonlinear regression within machine learning (ML), and surrogate modeling. GPs are analytically tractable and can be fully specified in terms of a prior mean function and a kernel, also known as a covariance function.

Keywords machine learning energy storage early lifetime prediction failure probability distribution Gaussian processes experimental stopping criterion distribution entropy domain-informed machine learning Introduction As the need for energy increases at an accelerating rate, there has been a global adoption of additional energy sources, such as solar and wind.

Keywords machine learning Bayesian optimization Gaussian process stochastic modeling parameter space efficient exploration uncertainty quantification energy storage battery failure modeling complex cycling conditions Introduction In response to the increasing energy demand, there has been a significant surge in the development of new energy storage technologies. Although solar and wind have great potential to increase energy supply, their intermittent nature hinders grid integration.