CSE 648
206-543-2969
pedrodcs.washington.edu
Areas of interest: 

Machine learning, artificial intelligence, data science

Research Interests

My main research interests are in the fields of machine learning and data mining. I'd like to make computers do more with less help from us, learn from experience, adapt effortlessly, and discover new knowledge. We need computers that reduce the information overload by extracting the important patterns from masses of data. This poses many deep and fascinating scientific problems: How can a computer decide autonomously which representation is best for target knowledge? How can it tell genuine regularities from chance occurrences? How can pre-existing knowledge be exploited? How can a computer learn with limited computational resources? How can learned results be made understandable by us?

My research addresses these and related questions. Research topics that I'm working on, or have recently worked on, include:

  • Learning concepts represented by sets of rules
  • Using examples as implicit definitions of concepts
  • Using probabilistic representations and analyses to address the uncertainty inherent in learning
  • Automating the process of selecting representations for concepts
  • Learning several models and combining them to improve accuracy and stability
  • Evaluating and selecting candidate models to avoid "overfitting" (i.e., to distinguish between genuine regularities and chance occurrences)
  • Learning models that can be easily understood by people
  • Using pre-existing knowledge to guide and improve learning
  • Developing knowledge discovery algorithms that run in linear or near-linear time, and so scale up to large databases
  • Using subsampling techniques to scale up pre-existing approaches
  • Developing algorithms that take into account the costs of decisions
  • Understanding the probabilistic properties and foundations of data mining algorithms
  • Developing techniques for mining semi-structured data sources (e.g., text, the Web)