Research Overview

Our research interest is statistical machine learning and data mining. We are particularly interested in modeling and analysis of unstructured data (such as image, function, shape, direction and text data), and modeling time lapsed unstructured data for understanding and controlling time-varying processes involving changes in unstructured data, with applications to data driven discovery of advanced materials and data driven monitoring and control of advanced manufacturing processes.

Our other research is integration of theoretical models and data models through surrogate modeling, model calibration and uncertainty quantification mainly because many time-varying processes we are studying are partly explained by some theorectical models up to a certain degree with uncertainty. Many theorectical models are described as partial differential equations, and a popular choice for a surrogate model and model calibration is Gaussian process. Our interest is on how to model Gaussian process with partial differential equations.

Recent News

  • August 31 2018 - Our paper with Daniel Apley was publisehd in Journal of Machine Learning Research (paper).
  • August 14 2018 - Ali's paper was publisehd in Technometrics (paper).
  • August 4 2018 - Xin Li received PhD degree. He will join Oak Ridge National Lab.
  • April 30 2018 - Vo's paper was published in Pattern Recognition (paper)
  • April 15 2018 - Our lab received funding from Air Force Office of Scientific Research - Dynamic Data Driven Application Systems Program
  • March 9 2018 - Xin's paper was published in Annals of Applied Statistics (paper)