Overview

Real-time control systems are pivotal in a myriad of sectors, including manufacturing, transportation, robotics, energy, and process systems, playing a crucial role in facilitating effective decision-making and coordination. The challenge lies in designing control systems that are both optimal and efficient, a task complicated by the inherent complexity and uncertainty of the processes involved. Bayesian Optimization (BO) emerges as a potent solution to these challenges. As a sophisticated black-box optimization technique, BO offers a data-efficient and potent strategy for optimizing control systems in real time. As such there is a growing interest on Bayesian optimization within the control community. 

The objective of this tutorial workshop is to provide participants with a comprehensive understanding of Bayesian optimization and its applications in real-time control and optimization. The workshop will cover the theoretical foundations of Bayesian optimization, including Gaussian processes and acquisition functions. Furthermore, the workshop will present cutting-edge research in Bayesian Optimization tailored for real-time control, encompassing constrained Bayesian optimization, safe Bayesian optimization, BO for high-dimensional and time-varying problems, and multi-agent Bayesian optimization. The workshop will also showcase the use of BO in a broad range of application areas including robotics, race cars, process control, digital health, and manufacturing systems. 

Target audience

This workshop is designed for researchers, practitioners, and graduate students in the fields of control engineering, optimization, and machine learning, offering them an opportunity to deepen their understanding and skills in these areas. No prior familiarity with Bayesian Optimization or Gaussian Processes is necessary for participation. Scheduled as a full-day event, the workshop will feature lectures on state-of-the-art algorithms and their applications in real-time control using Bayesian Optimization. Additionally, it will include practical, hands-on coding sessions. These sessions are aimed at helping participants establish a Bayesian Optimization code base, enhancing their practical skills and application knowledge.

New! Take home a practical code-base from our hands-on tutorial, that will help you get started with BO in your project!

Organizers: Yuning Jiang, Dinesh Krishnamoorthy, Wenjie Xu, Colin N. Jones

Contact: Yuning Jiang (yuning.jiang@ieee.org, yuning.jiang@epfl.ch)