This was part of Materials Informatics: Tutorials and Hands-On

Bayesian optimization for real-world experimental design

Yuxin Chen, University of Chicago

Friday, March 15, 2024

Effective black-box optimization is a widespread challenge in many real-world applications, including optimal experimental design in science and engineering (e.g., robotic control, molecular engineering, and drug discovery). Thus far, Bayesian optimization (BO) has been a state-of-the-art technique, and among the most popular approaches that empower these modern applications. This tutorial provides a comprehensive starting point for researchers and practitioners interested in the theory and practice of real-world active learning and Bayesian optimization.

In this tutorial, I will first lay out the basics of active learning and BO, including the generic problem setup and the fundamental algorithms. I will then survey a broad range of representative challenges that arise in real-world experimental design applications, along with recent advances for addressing such challenges. We will base our discussion on successful applications of modern BO algorithms that have had real-world impact.