An Active Archicture for Self-Learning

Abstract

Create a novel architecture that extracts events from the flow of time in an unsupervised manner and substitutes the conventional classifier by an external memory system that stores and organizes the extracted events to foster continuous growth in the memory store and continuous improvement of performance.

Document Details

Document Type
DoD Grant Award
Publication Date
Jul 15, 2019
Source ID
FA94531810039

Entities

People

  • José Príncipe

Organizations

  • Air Force Research Laboratory
  • Defense Advanced Research Projects Agency
  • University of Florida

Tags

Fields of Study

  • Computer science

Readers

  • Defense Technology Research and Development.
  • Neural Network Machine Learning.
  • Parallel and Distributed Computing.