Maturation of Determining the Limits of AI Robustness (MDLAR)

Abstract

MDLAR: Military Application: The Department of Defense (DoD): Is increasingly relying on machine learning (ML) systems for a variety of mission and support functions; Requires rapid adaptation to new situations, environments, and threats; Needs confidence that, when deploying ML systems, they will perform accurately; Needs to know when to wait for data to retrain/boost before their continued use. This one-year project focused on: Developing an approach for determining the robustness of artificial intelligence (AI) solutions; Building an application (prototype) to do just that within a narrow domain.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Nov 14, 2022
Accession Number
AD1180836

Entities

People

  • Michael D. Konrad

Organizations

  • Carnegie Mellon University

Tags

DTIC Thesaurus Topics

  • Algorithms
  • Artificial Intelligence
  • Copyrights
  • Department Of Defense
  • Engineering
  • Governments
  • Guarantees
  • Information Science
  • Machine Learning
  • Materials
  • Maturation
  • Military Applications
  • Software Development
  • Statistics
  • Transitions
  • Universities

Readers

  • Geospatial Intelligence and Artificial Intelligence Analytics
  • Military Mobilization and Reserve Forces Studies.
  • Systems Analysis and Design

Technology Areas

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