Skip to main navigation Skip to search Skip to main content

Learning from Hindsight Demonstrations

  • Harbin Institute of Technology

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Learning from demonstrations (LfD) is an important technique to help reinforcement learning (RL) boost the training process, especially in the case of sparse rewards. But a major obstacle is the acquisition of expert demonstrations, which is difficult or expensive to obtain in many cases. In this paper, we propose a unique method called Learning from Hindsight Demonstrations (LfHD) to automatically produce hindsight demonstrations, on which LfD can be performed and the cost of acquiring expert demonstrations is avoided. The produced demonstrations are comparable to those of experts at certain success rate. We also improve the LfD method to make better use of the produced demonstrations. Experiments show that our method can greatly improve the training efficiency compared to existing algorithms.

Original languageEnglish
Title of host publicationNeural Information Processing - 29th International Conference, ICONIP 2022, Proceedings
EditorsMohammad Tanveer, Sonali Agarwal, Seiichi Ozawa, Asif Ekbal, Adam Jatowt
PublisherSpringer Science and Business Media Deutschland GmbH
Pages480-491
Number of pages12
ISBN (Print)9789819916412
DOIs
StatePublished - 2023
Event29th International Conference on Neural Information Processing, ICONIP 2022 - Virtual, Online
Duration: 22 Nov 202226 Nov 2022

Publication series

NameCommunications in Computer and Information Science
Volume1792 CCIS
ISSN (Print)1865-0929
ISSN (Electronic)1865-0937

Conference

Conference29th International Conference on Neural Information Processing, ICONIP 2022
CityVirtual, Online
Period22/11/2226/11/22

Keywords

  • hindsight experience replay
  • learning from demonstrations
  • reinforcement learning

Fingerprint

Dive into the research topics of 'Learning from Hindsight Demonstrations'. Together they form a unique fingerprint.

Cite this