What is Your Metric Telling You? Evaluating Classifier Calibration under Different Definitions of Reliability

Abstract

An argument for uncertainty in machine learned models. This work: Evaluating classifiers for context-specific calibration. How do you evaluate your classifier for it's ability to accurately express uncertainty? First, you need to define how to interpret the confidence outputs of classifiers.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 2021
Accession Number
AD1150239

Entities

People

  • Eric T. Heim

Organizations

  • Carnegie Mellon University

Tags

Communities of Interest

  • Autonomy

DTIC Thesaurus Topics

  • Artificial Intelligence
  • Calibration
  • Department Of Defense
  • Engineering
  • Governments
  • Guarantees
  • Learning
  • Literature
  • Machine Learning
  • Materials
  • Probability
  • Reliability
  • Sampling
  • Software Development
  • Uncertainty
  • Universities

Readers

  • Aerospace Test and Evaluation
  • Neural Network Machine Learning.
  • Systems Analysis and Design

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • AI & ML - Neural Networks