Sample-Efficient Reinforcement Learning with Symmetry-Guided Demonstrations for Robotic Manipulation
This work addresses sample efficiency for robotic manipulation tasks, offering a domain-specific improvement that is incremental by building on existing RL methods with demonstrations.
The paper tackles the problem of low sample efficiency in reinforcement learning for robotic manipulation by introducing Demo-EASE, a framework that uses symmetry-guided demonstrations and a dual-buffer architecture, resulting in significantly accelerated convergence and improved final performance in simulation experiments with a Kinova Gen3 robot.
Reinforcement learning (RL) suffers from low sample efficiency, particularly in high-dimensional continuous state-action spaces of complex robotic manipulation tasks. RL performance can improve by leveraging prior knowledge, even when demonstrations are limited and collected from simplified environments. To address this, we define General Abstract Symmetry (GAS) for aggregating demonstrations from symmetrical abstract partitions of the robot environment. We introduce Demo-EASE, a novel training framework using a dual-buffer architecture that stores both demonstrations and RL-generated experiences. Demo-EASE improves sample efficiency through symmetry-guided demonstrations and behavior cloning, enabling strong initialization and balanced exploration-exploitation. Demo-EASE is compatible with both on-policy and off-policy RL algorithms, supporting various training regimes. We evaluate our framework in three simulation experiments using a Kinova Gen3 robot with joint-space control within PyBullet. Our results show that Demo-EASE significantly accelerates convergence and improves final performance compared to standard RL baselines, demonstrating its potential for efficient real-world robotic manipulation learning.