Concepedia

Bootstrapped Exploration

Bootstrapped Exploration is a reinforcement learning exploration strategy that belongs to the class of uncertainty-aware methods, distinguished by its use of bootstrapped sampling from historical data or policy outputs to generate diverse and adaptive exploration behavior without requiring explicit environmental interaction.

3

Publications

544

Citations

7

Authors

4

Institutions

Publications per year

2017–2026

1

Parents

Publications

3

1–3 of 3

Authors

7

  1. IO
    Ian Osband

    Stanford University, United States

    2 publications · 528 citations

    2

    h-index

  2. BV
    Benjamin Van Roy

    Stanford University, United States

    2 publications · 528 citations

    2

    h-index

  3. AP
    Alexander Pritzel

    Google DeepMind (United Kingdom), United Kingdom

    1 publications · 468 citations

    1

    h-index

  4. CB
    Charles Blundell

    Google DeepMind (United Kingdom), United Kingdom

    1 publications · 468 citations

    1

    h-index

  5. WH
    Wenzhen Huang

    University of Massachusetts Dartmouth, United States

    1 publications · 16 citations

    1

    h-index

1–5 of 7

Institutions

4

  1. Google DeepMind (United Kingdom)

    London, United Kingdom

    1 h-index

    3 publications

    1.4K citations

  2. Chinese Academy of Sciences

    Beijing, China

    1 h-index

    3 publications

    48 citations

  3. Stanford University

    Stanford, United States

    1 h-index

    2 publications

    936 citations

  4. Google (United States)

    Mountain View, United States

    1 h-index

    3 publications

    1.4K citations

1–4 of 4

Venues

  1. arXiv (Cornell University)

    Repository · 2 publications · 528 citations

    2

    h-index

  2. Proceedings of the AAAI Conference on Artificial Intelligence

    Conference · 1 publications · 16 citations

    1

    h-index

1–2