Interactive Policy Refinement

Interactive Policy Refinement is a machine learning framework within human-robot interaction that enables autonomous agents to iteratively improve their decision-making policies through direct, real-time feedback from human users, distinguishing it from passive or offline policy optimization methods by emphasizing bidirectional interaction and adaptive learning during task execution.

3

Publications

139

Citations

15

Authors

6

Institutions

Publications per year

2017–2026

2

Authors

15

  1. KG
    Kobi Gal

    Ben-Gurion University of the Negev, Israel

    1 publications · 15 citations

    1

    h-index

  2. EK
    Ece Kamar

    Microsoft (United States), United States

    1 publications · 15 citations

    1

    h-index

  3. TD
    Trevor Darrell

    University of California, Berkeley, United States

    1 publications · 99 citations

    1

    h-index

  4. EH
    Eric Horvitz

    Microsoft (United States), United States

    1 publications · 15 citations

    1

    h-index

  5. AS
    Avi Segal

    Ben-Gurion University of the Negev, Israel

    1 publications · 15 citations

    1

    h-index

1–5 of 15

Venues

  1. arXiv (Cornell University)

    Repository · 1 publications · 99 citations

    1

    h-index

  2. Proceedings of the AAAI Conference on Artificial Intelligence

    Conference · 1 publications · 15 citations

    1

    h-index

  3. Springer proceedings in advanced robotics

    SPAR · Book series · 1 publications · 12 citations

1–3