Where-to-Learn: Analytical Policy Gradient Directed Exploration for On-Policy Robotic Reinforcement Learning

TL;DR

This paper introduces a novel exploration method for on-policy robotic reinforcement learning that uses analytical policy gradients from a differentiable dynamics model to guide the agent towards high-reward states, improving learning efficiency.

Contribution

It presents a new directed exploration approach leveraging physics-guided, analytical policy gradients to enhance on-policy RL in robotics, unlike entropy-based methods.

Findings

Accelerated policy learning in robotic control tasks.

Effective steering towards high-reward regions.

Improved sample efficiency over traditional exploration methods.

Abstract

On-policy reinforcement learning (RL) algorithms have demonstrated great potential in robotic control, where effective exploration is crucial for efficient and high-quality policy learning. However, how to encourage the agent to explore the better trajectories efficiently remains a challenge. Most existing methods incentivize exploration by maximizing the policy entropy or encouraging novel state visiting regardless of the potential state value. We propose a new form of directed exploration that uses analytical policy gradients from a differentiable dynamics model to inject task-aware, physics-guided guidance, thereby steering the agent towards high-reward regions for accelerated and more effective policy learning.