HG-DAgger: Interactive imitation learning with human experts

Michael Kelly, Chelsea Sidrane, Katherine Driggs-Campbell, Mykel J. Kochenderfer

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Imitation learning has proven to be useful for many real-world problems, but approaches such as behavioral cloning suffer from data mismatch and compounding error issues. One attempt to address these limitations is the DAgger algorithm, which uses the state distribution induced by the novice to sample corrective actions from the expert. Such sampling schemes, however, require the expert to provide action labels without being fully in control of the system. This can decrease safety and, when using humans as experts, is likely to degrade the quality of the collected labels due to perceived actuator lag. In this work, we propose HG-DAgger, a variant of DAgger that is more suitable for interactive imitation learning from human experts in real-world systems. In addition to training a novice policy, HG-DAgger also learns a safety threshold for a model-uncertainty-based risk metric that can be used to predict the performance of the fully trained novice in different regions of the state space. We evaluate our method on both a simulated and real-world autonomous driving task, and demonstrate improved performance over both DAgger and behavioral cloning.

Original languageEnglish (US)
Title of host publication2019 International Conference on Robotics and Automation, ICRA 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages8077-8083
Number of pages7
ISBN (Electronic)9781538660263
DOIs
StatePublished - May 2019
Event2019 International Conference on Robotics and Automation, ICRA 2019 - Montreal, Canada
Duration: May 20 2019May 24 2019

Publication series

NameProceedings - IEEE International Conference on Robotics and Automation
Volume2019-May
ISSN (Print)1050-4729

Conference

Conference2019 International Conference on Robotics and Automation, ICRA 2019
CountryCanada
CityMontreal
Period5/20/195/24/19

ASJC Scopus subject areas

  • Software
  • Control and Systems Engineering
  • Artificial Intelligence
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'HG-DAgger: Interactive imitation learning with human experts'. Together they form a unique fingerprint.

  • Cite this

    Kelly, M., Sidrane, C., Driggs-Campbell, K., & Kochenderfer, M. J. (2019). HG-DAgger: Interactive imitation learning with human experts. In 2019 International Conference on Robotics and Automation, ICRA 2019 (pp. 8077-8083). [8793698] (Proceedings - IEEE International Conference on Robotics and Automation; Vol. 2019-May). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICRA.2019.8793698