Human-in-the-Loop Bayesian Optimization for ALL: Practical Applications in Human-Computer Interaction Challenges

Bayesian optimization provides a powerful, sampling-efficient approach to designing user interfaces that meet design objectives. It builds a probabilistic model of how well a design variant fulfills the objectives and then selects the next most promising variant to improve over a few iterations. When paired with human feedback on designs, such as subjective assessments of trust, usability, or cognitive load, it enables a human-in-the-loop workflow that accelerates interactive design and evaluation. In this tutorial, we blend theoretical foundations with practical, hands‑on demonstrations to equip attendees with the skills needed to deploy both single- and multi-objective Bayesian optimization pipelines in human-computer interaction contexts. Through live coding, integration examples in Unity, and open discussion, attendees will leave ready to apply these methods in their own research and product development.