Interactive Learning from Sparse and Diverse Feedback

Abstract

Despite the great success of machine learning, most learning algorithms remain primarily non-interactive. Human learning, on the other hand, is highly interactive. As learners we are inquisitive - we ask questions about the data shown to us and employ feedback to decide what questions would yield answers that are most informative for the learning task at hand. In this project, we developed a suite of data-efficient interactive machine learning algorithms that employ judicious choice of what data to collect. This includes development of new algorithms as well as integration of algorithms developed in PIs' prior works.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Mar 01, 2020
Accession Number
AD1092496

Entities

People

  • Aarti Singh
  • Artur Dubrawski
  • Barnabas Poczos

Organizations

  • Carnegie Mellon University

Tags

Communities of Interest

  • Autonomy

DTIC Thesaurus Topics

  • Algorithms
  • Artificial Intelligence Computing
  • Artificial Intelligence Software
  • Compressed Sensing
  • Convolutional Neural Networks
  • Deep Learning
  • Dictionaries
  • Dimensionality Reduction
  • Experimental Design
  • Gaussian Processes
  • Information Processing
  • Information Science
  • Information Systems
  • Machine Learning
  • Neural Networks
  • Signal Processing
  • Unsupervised Machine Learning

Fields of Study

  • Computer science

Readers

  • Neural Network Machine Learning.
  • Systems Analysis and Design

Technology Areas

  • AI & ML