Perspectives on Bayesian Optimization for HCI

Bjørn Sand Jensen, Jens Brehm Bagger Nielsen, Jan Larsen

AbstractIn this position paper we discuss optimization in the HCI domain based on our experiences with Bayesian methods for modeling and optimization of audio systems, including challenges related to evaluating, designing, and optimizing such interfaces. We outline and demonstrate how a combined Bayesian modeling and optimization approach provides a flexible framework for integrating various user and content attributes, while also supporting model-based optimization of HCI systems. Finally, we discuss current and future research direction and applications, such as inferring user needs and optimizing interfaces for computer assisted teaching.
KeywordsBayes, Optimization, Gaussian process priors, Modeling
TypeConference paper [With referee]
ConferenceCHI'15 Workshop on Principles, Techniques and Perspectives on Optimization and HCI
Year2015    Month April
NoteSeoul, South Korea, April 19
Electronic version(s)[pdf]
Publication link
BibTeX data [bibtex]
IMM Group(s)Intelligent Signal Processing

Back  ::  IMM Publications