As one of the most exciting multi-disciplinary research fields,biomedical informatics has seen dramatic growth over the past decade, especially with respect to newly available data. This rapidly-growing resource has posted many interesting new challenges for machine learning since biomedical data are often noisy, complex, relationally structured and highly diverse. In this talk, I will present a number of machine learning approaches we have proposed, covering topics like semi-supervised learning, multitask learning, feature learning, dimension reduction, deep learning, etc, to handle different types of data complexities that are urgent to be addressed in biomedical domain.
Yanjun obtained her Ph.D. degree from School of Computer Science at Carnegie Mellon University in 2008 and received her Bachelor degree with high honors from Computer Science Department at Tsinghua University, Beijing. She co-chaired the "NIPS Machine Learning in Computational Biology" workshop from 2009 to 2011.