In industrial R&D and process improvement we experiment to understand complex, multifactor systems. There has been tremendous interest in applying "Bayesian optimization" or "Active Learning" to efficiently innovate new products and processes. These approaches to sequential experimentation promise greater speed while being more approachable to scientists and engineers. Generalizing Bayesian optimization (Bayes opt.) to real world complex problems involving multiple responses has proven challenging because in its standard formulation the Bayes opt. approach is inherently limited to a single response. In this webinar we review the basics of Gaussian Process regression modeling and the standard approach to Bayes opt. We then introduce the generalization to multiple responses via the Bayesian Desirability framework. We will demonstrate the efficiency and approachability of the technique using new capabilities in JMP Pro 19.
Materials to try these methods yourself in JMP Pro version 19.0 are attached. If you are a student or academic researcher, you can get free access to JMP Student Edition here.
Bayesian Optimization Journal (Participants).jrn
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.