Assistant Professor in Applied Mathematics

Data Science Fellow

Specializations
Applied Mathematics, Machine Learning, Statistics

Biography

Washington Research Foundation Data Science Professor in Applied Mathematics Data Science Fellow, University of Washington eScience Institute

Research goals:

Aravkin’s work is focused on design and analysis of custom formulations and algorithms to extract meaningful information from noisy data. This is a crucial problem in a broad range of areas, including classification and prediction, recommender systems, time series, seismic imaging, tracking and navigation, and diagnostic medicine, to name a few. Inference in all these contexts is made more challenging by overwhelming data size, presence of confounding signals, and complexity of our models of physical phenomena. Aravkin tries to design formulations that use robust statistical modeling and prior information about the problem to obtain useful solutions in difficult situations.

Once formulated, we need fast reliable algorithms to fit models. Aravkin is particularly interested in algorithms that exploit special structure, also making use of randomized techniques to probe large data volumes. He loves to work across a wide range of topics, from general development of optimization theory and algorithms, to statistical and physical modeling, and finally to customized algorithms for particular application domains.

Optimization and statistics connect to a wide range of application domains, but these connections must be found. For Aravkin, the key to interdisciplinary work is to find enthusiastic collaborators, who help him learn about new fields, understand open problems, and start to see how he can help make an impact. The interdisciplinary environment of the eScience Institute provides fantastic support for this work, including dedicated space and resources to encourage researchers from a range of areas to come together and explore applications that can benefit from big data techniques. Aravkin is thrilled to be part of the eScience community at the UW.

Education history:

  • D. Mathematics, University of Washington, 2010
  • S. Statistics, University of Washington, 2010
  • Sc. Mathematics and Computer Science, University of Washington, 2004

Previous appointments:

  • Research Staff Member, IBM T.J. Watson Research Center, 2013-15
  • Adjunct Professor, Computer Science and IEOR, Columbia, 2014-15
  • Postdoctoral Research Fellow, Computer Science and Earth & Ocean Sciences, UBC, 2010-12