Data of growing complexity come from multiple sources with multiple aspects. For example, brain imaging, genetic data, and clinical records are collected to study the relationships between brain structure and function; and movie titles, genres, directors and ratings are available online, useful for modeling multiple aspects of users' preference. These data present us with unprecedented opportunities to integrate them with predictive models to extract complex relationships among natural and man-made objects. The objective of this project is to harness these opportunities and address computational and statistical challenges associated with them.
Reference:
Y Qi, F. Yan, EigenNet: A Bayesian hybrid of generative and conditional models for sparse learning, in Advances in Neural Information Processing Systems, 2011.
Supported by National Science Foundation