Cost-Aware defense of Sensors-to-Decisions System against Malicious Data Attacks

Abstract

The crucial function of sensors-to-decision (S2D) systems makes them appealing targets for cyber warfare. In this proposal, we address the problem of data attacks on an S2D system wherein an adversary falsifies part of data from sensors to perturb machine learning decisions of the system. We will develop a novel attack-cost-aware countermeasure as well as attack information visualization techniques which take into account the vulnerability characteristics of S2D systems of our interest.

Document Details

Document Type
DoD Grant Award
Publication Date
Jul 29, 2020
Source ID
N001741910009

Entities

People

  • Jinsub Kim

Organizations

  • Oregon State University
  • United States Navy

Tags

Fields of Study

  • Computer science

Readers

  • Aviation Science / Aeronautics.
  • Distributed Systems and Data Platform Development
  • Irregular Warfare and Special Operations Cyberspace Operations against Adversarial Threats.

Technology Areas

  • AI & ML
  • AI & ML - DoD AI Strategy
  • Cyber