SUCCESS: Self-assessment and Understanding of Competence and Conditions to Ensure System Success

Abstract

The core focus of the proposed research is to develop new knowledge and techniques for machine self-assessment of pro ciency. We describe this challenge as a taxonomy of two intersecting dimensions: Time and evels of Self-Assessment. For the former, these approaches need to work a priori, in situ, and post hoc in order to support e ective autonomy and utilization by human partners and managers. For the latter, self-assessment can range from simple detection of pro ciency up through evaluation, explanation, and prediction. The bulk of the proposed e ort will focus on the exploration, testing, and development of promising techniques for calculating pro ciency self-assessment. Where appropriate, we plan to draw inspiration from existing cognitive and statistical models of human pro ciency from psychology and other elds. While computers and humans have very di erent internal cognitive processes, there are interesting approaches that can be approximated.

Document Details

Document Type
DoD Grant Award
Publication Date
Jul 26, 2018
Source ID
N000141812503

Entities

People

  • Aaron Steinfeld

Organizations

  • Massachusetts Institute of Technology
  • Office of Naval Research
  • United States Navy

Tags

Readers

  • Distributed Systems and Data Platform Development
  • Regression Analysis.
  • Team-Based Human-Centered Cognitive Task Decision Making and Information Performance.