Discovering Common Sense from Video, Images, Text and Knowledge Bases

Abstract

The scope of this effort is to automatically learn a common sense repository (CSR) of knowledge that can be applied to diverse problems. USC will do this by embedding spaces based on deep neural nets (DNNs). Embedding spaces have recently proven effective on tasks that leverage common sense. The immediate impact will be more effective artificial intelligence (AI) applications, since common sense applications learn via their embedding space, they will be domain-independent but orientable for a particular purpose. The research will both reduce the need for domain-specific training and increase accuracy of AI applications.

Document Details

Document Type
DoD Grant Award
Publication Date
Jan 13, 2022
Source ID
N660011924032

Entities

People

  • Ralph Weischedel

Organizations

  • Defense Advanced Research Projects Agency
  • Naval Information Warfare Center Pacific
  • University of Southern California

Tags

Fields of Study

  • Computer science

Readers

  • Artificial Intelligence
  • Neural Network Machine Learning.

Technology Areas

  • AI & ML
  • AI & ML - DoD AI Strategy
  • AI & ML - Neural Networks
  • Space