Statistical Machine Learning with Restrictions on Shape, Structure, and Computation

Abstract

Statistical Machine Learning with Restrictions on Shape, Structure and Computation The research carried out in this project is targeted at advancing the mathematical foundations of data science, by developing theory and methodology for problems that have not been studied in traditional statistical and machine learning approaches. The research will develop models, theories and algorithms that can help scientists and engineers conduct more effective data analysis. The focus of the project is on statistical machine learning problems that impose restrictions on the computational resources consumed, or on the shape of estimators such as convexity and monotonicity. In addition, mathematical problems in graph-structured signal processing will be studied. A fourth area of investigation will be the use of statistical learning in optimization and scientific computing. The potential relevance to Navy interests involves any domain where data processing is central, including sensing, tracking, communication and information management tasks.

Document Details

Document Type
DoD Grant Award
Publication Date
Aug 12, 2016
Source ID
N000141512379

Entities

People

  • John D. Lafferty

Organizations

  • Office of Naval Research
  • United States Navy
  • University of Chicago

Tags

Readers

  • Neural Network Machine Learning.
  • Research Science/Academic Research
  • Statistical inference.

Technology Areas

  • AI & ML
  • AI & ML - Machine Learning Algorithms
  • AI & ML - Neural Networks