I work in artificial intelligence and machine learning, focusing on statistical methods for learning from data and on approximate inference techniques for graphical models. Applications of my work include data mining and information fusion in sensor networks, computer vision and image processing, and computational biology.

Graphical models are used to organize and structure probability distributions over large systems, and enable efficient approximate or exact reasoning. My group balances developing theoretical and algorithmic advances with applications to the real-world systems of our collaborators.

Algorithms. One of our main focuses is on finding maxima or computing probabilities using variational methods, including the family of belief propagation (BP) message-passing algorithms. Our contributions include analyzing the convergence and accuracy properties of BP, developing new BP-like bounds, extending BP techniques to continuous valued systems, and improving the efficiency of "adaptive" or incremental inference.

Applications. We have applied our algorithms to a wide variety of problems, including tracking and understanding data from sensor networks, efficient representations for large text corpora, computer vision and image processing, and gene expression data in biology.

Qiang Liu has received a 2011 Microsoft Research Fellowship award.

Current:

• Ozgur Sumer
• Andrew Frank
• Qiang Liu
• David Keator
• Sholeh Forouzan

Graduated:

• Jonathan Hutchins (PhD, 2010)
• Sidharth Shekhar (MS, 2009)
• Priya Venkateshan (MS, 2011)

• UCI's Center for Machine Learning and our AI/ML seminar series
• UCI Machine Learning Dataset Repository

We gratefully acknowledge support for our current and recent research from the National Science Foundation, Microsoft Research, the National Institute of Health, NIAMS, and UCI's Center for Complex Biological Systems.