Making Bayesian Optimization Practical in the Context of High Dimensional, Highly Expensive, Black­Box Functions