This year, we released version 1.0 of Ax, an open source adaptive experimentation platform that leverages machine learning to guide and automate the experimentation process. Ax employs Bayesian optimization to enable researchers and developers to conduct efficient experiments, identifying optimal configurations to optimize their systems and processes.
In conjunction with this major release, we published a paper titled, “Ax: A Platform for Adaptive Experimentation” that explores Ax’s core architecture, provides a deeper explanation of the methodology powering the optimization, and compares Ax’s performance against other black-box optimization libraries.