Reinforcement Learning for Robotics and Control with Active Uncertainty Reduction

TL;DR

This paper introduces an active uncertainty reduction approach that combines model-based and model-free reinforcement learning to improve data efficiency and robustness in robotic control tasks.

Contribution

It presents a novel method for active uncertainty management using virtual environments and adaptive sampling, enhancing learning efficiency and system modeling.

Findings

Better modeling of complex system dynamics compared to existing methods.

Efficient uncertainty management improves policy learning.

Applicable to various reinforcement learning problems in OpenAI gym.

Abstract

Model-free reinforcement learning based methods such as Proximal Policy Optimization, or Q-learning typically require thousands of interactions with the environment to approximate the optimum controller which may not always be feasible in robotics due to safety and time consumption. Model-based methods such as PILCO or BlackDrops, while data-efficient, provide solutions with limited robustness and complexity. To address this tradeoff, we introduce active uncertainty reduction-based virtual environments, which are formed through limited trials conducted in the original environment. We provide an efficient method for uncertainty management, which is used as a metric for self-improvement by identification of the points with maximum expected improvement through adaptive sampling. Capturing the uncertainty also allows for better mimicking of the reward responses of the original system. Our approach enables the use of complex policy structures and reward functions through a unique combination of model-based and model-free methods, while still retaining the data efficiency. We demonstrate the validity of our method on several classic reinforcement learning problems in OpenAI gym. We prove that our approach offers a better modeling capacity for complex system dynamics as compared to established methods.