Mark Craven's Research Group

• identifying gene-regulatory elements in genomic sequences,
• uncovering and modeling networks of interactions among genes and other cellular components,
• modeling, classifying and aligning temporal gene-expression responses induced by toxicants,
• annotating high-throughput biological experiments,
• extracting structured information from the scientific literature.

• Learning expressive models for sequence data, including models that represent overlapping elements, complex configurations of sequence elements, and real-valued outputs conditioned on sequence features.
• Learning in problem domains where most instances are either unlabeled or labeled at a coarse level of granularity. We are developing novel methods for active learning and multiple-instance learning.
• Learning models that represent high-dimensional, sparsely-sampled time series.

Current Postdocs

• Irene Ong
• Andreas Vlachos

Current PhD Students

• Debbie Chasman
• Deborah Muganda
• Anna Rissman (co-advised by Michael Gould)

Graduated PhD Students

• Dave Andrzejewski, PhD 2010 (co-advised by Xiaojin Zhu)
  thesis: Incorporating Domain Knowledge in Latent Topic Models
  current position: Postdoctoral Researcher, Lawrence Livermore National Laboratory

• Joe Bockhorst, PhD 2005
  thesis: Machine Learning Methods for Discovering Regulatory Elements in Bacterial Genomes
  current position: Assistant Professor, University of Wisconsin-Milwaukee

• Keith Noto, PhD 2007
  thesis: Learning Expressive Computational Models of Gene Regulatory Sequences and Responses
  current position: Postdoctoral Researcher, Tufts University

• Yue Pan, PhD 2009
  thesis: Inferring Mechanism-Based Gene Regulatory Network Models from Expression and Sequence Data
  current position: Research Scientist, Facebook

• Soumya Ray, PhD 2005 (co-advised by David Page)
  thesis: Learning from Data with Complex Interactions and Ambiguous Labels
  current position: Assistant Professor, Case Western Reserve University

• Burr Settles, PhD 2008
  thesis: Curious Machines: Active Learning with Structured Instances
  current position: Postdoctoral Fellow, Carnegie Mellon University

• Adam Smith, PhD 2009
  thesis: Classification and Alignment of Gene-Expression Time-Series Data
  current position: Postdoctoral Researcher, Oregon Health Sciences University