Robust Command and Control under Adversarially Perturbed Data

Abstract

All command and control (C2) decisions are founded upon information. This information may be quantitative or qualitative and in structured or unstructured forms. Low quality information can adversely impact a commander's C2 decisions for their forces. As such, influence over an adversary's perceived information and data has been a central tenet of national competition throughout history. This is evident from the ancient writings of Sun Tzu to more contemporary examples in World War II (e.g., Operation Fortitude). The emerging Joint All Domain Command and Control (JADC2) construct entails a highly networked infrastructure heavily reliant on machine learning for data fusion and synthesis. Direct manipulation of digital data (e.g., airborne sensor output, social media posts) can significantly alter the performance of such algorithms and therefore constitutes a serious threat to an emerging friendly center of gravity. This research project will focus on the following areas: • Develop a general statistical framework, called Adversarial Statistical Decision Theory (ASDT), that encompasses standard inference and decision tasks (point and interval estimation, hypothesis testing, forecasting and decision support) when the underlying data itself is subject to attack from intelligent adversaries. • Implement the framework based on the development of computationally efficient algo­rithms for its real-time application. • While the generality of this research allows application to a broad array of C2 ac­tivities, we will develop methods and software specifically tailored to the information environment based on recent United States Air Force needs. • Besides this specific application, we shall also compile a list of other operational con­texts where our framework and concepts would be relevant.

Document Details

Document Type
DoD Grant Award
Publication Date
Jan 21, 2022
Source ID
FA86552117042XX0

Entities

People

  • David Rios

Organizations

  • Air Force Office of Scientific Research
  • Spanish National Research Council
  • United States Air Force

Tags

Fields of Study

  • Computer science

Readers

  • Military History / Militaries and War Studies
  • Neural Network Machine Learning.
  • Theoretical Analysis.

Technology Areas

  • AI & ML
  • AI & ML - DoD AI Strategy
  • Fully Networked C3
  • Fully Networked C3 - Command and Control