Episodic self-imitation learning with hindsight

Tianhong Dai*, Hengyan Liu, Anil Anthony Bharath

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

11 Citations (Scopus)

Abstract

Episodic self-imitation learning, a novel self-imitation algorithm with a trajectory selection module and an adaptive loss function, is proposed to speed up reinforcement learning. Compared to the original self-imitation learning algorithm, which samples good state–action pairs from the experience replay buffer, our agent leverages entire episodes with hindsight to aid self-imitation learning. A selection module is introduced to filter uninformative samples from each episode of the update. The proposed method overcomes the limitations of the standard self-imitation learning algorithm, a transitions-based method which performs poorly in handling continuous control environments with sparse rewards. From the experiments, episodic self-imitation learning is shown to perform better than baseline on-policy algorithms, achieving comparable performance to state-of-the-art off-policy algorithms in several simulated robot control tasks. The trajectory selection module is shown to prevent the agent learning undesirable hindsight experiences. With the capability of solving sparse reward problems in continuous control settings, episodic self-imitation learning has the potential to be applied to real-world problems that have continuous action spaces, such as robot guidance and manipulation.

Original languageEnglish
Article number1742
Pages (from-to)1-18
Number of pages18
JournalElectronics (Switzerland)
Volume9
Issue number10
DOIs
Publication statusPublished - 2020
Externally publishedYes

Keywords

  • Deep reinforcement learning
  • Exploration
  • Hindsight experience replay
  • Imitation learning

Fingerprint

Dive into the research topics of 'Episodic self-imitation learning with hindsight'. Together they form a unique fingerprint.

Cite this