· Contributors · Organizations · Search
User Interfaces for High-Dimensional Design Problems: From Theories to Implementations
Artificial Intelligence/Machine Learning
This session WILL be recorded.
Research & Education
DescriptionWe introduce techniques based on Bayesian optimization (BO) and differential subspace search (DSS) for building tractable user interfaces for design problems involving a large number of parameters. BO and DSS serve to extract much lower-dimensional yet meaningful subspaces that can be effectively used to create slider or gallery interfaces.
Prerequisites The audience is expected to have a basic understanding of linear algebra, calculus, and statistics.
Intended Audience Developers of design tools for general problems, including image editing, shape editing, motion editing, and sound editing. Developers of design tools using generative models from deep learning. Researchers in the topics related to user interfaces, machine learning, statistics, etc.