Storage, Representation, and Manipulation of Distribution Functions

Abstract

The Recipient will seek potential methods to represent, store, and manipulate multi-dimensional probability density functions that do not have a simple closed-form solution. The aim of the proposed research is to accurately estimate the probability of rare events from measurements, while considering samples of multi-dimensional random variables distributed in some unknown, correlated and complicated manner. The Recipient will also use multi-dimensional probability density function (PDF) to compute the probability of events of interest, and then update when new information arrives.

Document Details

Document Type
DoD Grant Award
Publication Date
May 17, 2017
Source ID
FA87501710113

Entities

People

  • Rajit Manohar

Organizations

  • Defense Advanced Research Projects Agency
  • Rome Laboratory
  • Yale University

Tags

Readers

  • Computational Modeling and Simulation
  • Statistical inference.