Episodic Self-Imitation Learning with Hindsight

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.

References in corpus (14)

• Continuous control with deep reinforcement learning
• A Brief Survey of Deep Reinforcement Learning
• End to End Learning for Self-Driving Cars
• Prioritized Experience Replay
• Dota 2 with Large Scale Deep Reinforcement Learning
• Leveraging Demonstrations for Deep Reinforcement Learning on Robotics Problems with Sparse Rewards
• Deep Q-learning from Demonstrations
• Multi-Goal Reinforcement Learning: Challenging Robotics Environments and Request for Research
• Reverse Curriculum Generation for Reinforcement Learning
• Reinforcement Learning from Imperfect Demonstrations
• Task-Embedded Control Networks for Few-Shot Imitation Learning
• Energy-Based Hindsight Experience Prioritization
• Learning Self-Imitating Diverse Policies
• Hindsight Generative Adversarial Imitation Learning

Cited by in corpus (1)

• Imaginary Hindsight Experience Replay: Curious Model-based Learning for Sparse Reward Tasks