NEARL: Non-Explicit Action Reinforcement Learning for Robotic Control

Abstract

Traditionally, reinforcement learning methods predict the next action based on the current state. However, in many situations, directly applying actions to control systems or robots is dangerous and may lead to unexpected behaviors because action is rather low-level. In this paper, we propose a novel hierarchical reinforcement learning framework without explicit action. Our meta policy tries to manipulate the next optimal state and actual action is produced by the inverse dynamics model. To stabilize the training process, we integrate adversarial learning and information bottleneck into our framework. Under our framework, widely available state-only demonstrations can be exploited effectively for imitation learning. Also, prior knowledge and constraints can be applied to meta policy. We test our algorithm in simulation tasks and its combination with imitation learning. The experimental results show the reliability and robustness of our algorithms.

References in corpus (10)

• End to End Learning for Self-Driving Cars
• Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm
• DeepMimic: Example-Guided Deep Reinforcement Learning of Physics-Based Character Skills
• Solving Rubik's Cube with a Robot Hand
• Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Models
• Learning human behaviors from motion capture by adversarial imitation
• Variational Discriminator Bottleneck: Improving Imitation Learning, Inverse RL, and GANs by Constraining Information Flow
• Constrained Policy Optimization
• Generative Adversarial Imitation from Observation
• Imitation Learning from Observations by Minimizing Inverse Dynamics Disagreement