Bayesian Optimization is an iterative optimization algorithm used to find the most optimal element like any other optimization algorithm however, where it shines in comparison to others is when the criterion or objective function is a black box function. A black box function indicates the prospect of a lack of an analytic expression or known mathematical formula for the objective function and/or details about other properties for example, knowing if the derivative exists for the function so as to make use of other optimization techniques such as Stochastic Gradient Descent.

I demonstrated how I developed and applied the Bayesian Optimization algorithm to various experiments including one making use of Event Tracing for Windows (ETW) profiles to obtain the best parameters and highlight how I made use of F#'s functional features to implement both the algorithm and the infrastructure associated with analyzing the results.