Pareto Optimal Layouts for Adaptive Mixed Reality
Published in Extended Abstracts of the 2023 CHI Conference on Human Factors in Computing Systems (CHI EA '23), 2023
Recommended citation: Christoph Albert Johns, João Marcelo Evangelista Belo, Clemens Nylandsted Klokmose, and Ken Pfeuffer. 2023. Pareto Optimal Layouts for Adaptive Mixed Reality. In Extended Abstracts of the 2023 CHI Conference on Human Factors in Computing Systems (CHI EA '23), April 23--28, 2023, Hamburg, Germany. ACM, New York, NY, USA 7 Pages. https://doi.org/10.1145/3544549.3585732 https://doi.org/10.1145/3544549.3585732
Adaptive mixed reality applications adjust their user interfaces based on the context in which they are used to provide a smooth experience for different users and environments. This involves carefully positioning UI elements, which can be challenging due to the many possible placements and need to balance competing goals. Current approaches employ global criterion optimization methods like weighted sums, which can be difficult to use, inflexible, and might not find preferred solutions. This can prevent the adaptations from meeting end-user expectations. We suggest using online multi-objective optimization methods which generate a set of Pareto optimal adaptation proposals, giving users more control and adding flexibility to the computational decision-making. We explore the feasibility of our approach by generating adaptations for a basic synthetic example and discuss relevant dimensions for a formal evaluation with end-users, including the choice of algorithm, decomposition technique, and objectives.