Curriculum Is More Influential Than Haptic Information During Reinforcement Learning of Object Manipulation Against Gravity

TL;DR

This study shows that curriculum learning significantly influences dexterous object manipulation in reinforcement learning, and tactile feedback is less critical than previously thought, with successful learning achieved without it.

Contribution

It introduces a novel curriculum-based learning rate scheduler and demonstrates that tactile feedback is not essential for learning complex manipulation tasks.

Findings

Curriculum learning biases feature acquisition in manipulation tasks.

Successful learning occurs without tactile feedback, challenging traditional views.

The curriculum-based scheduler accelerates convergence to higher rewards.

Abstract

Learning to lift and rotate objects with the fingertips is necessary for autonomous in-hand dexterous manipulation. In our study, we explore the impact of various factors on successful learning strategies for this task. Specifically, we investigate the role of curriculum learning and haptic feedback in enabling the learning of dexterous manipulation. Using model-free Reinforcement Learning, we compare different curricula and two haptic information modalities (No-tactile vs. 3D-force sensing) for lifting and rotating a ball against gravity with a three-fingered simulated robotic hand with no visual input. Note that our best results were obtained when we used a novel curriculum-based learning rate scheduler, which adjusts the linearly-decaying learning rate when the reward is changed as it accelerates convergence to higher rewards. Our findings demonstrate that the choice of curriculum greatly biases the acquisition of different features of dexterous manipulation. Surprisingly, successful learning can be achieved even in the absence of tactile feedback, challenging conventional assumptions about the necessity of haptic information for dexterous manipulation tasks. We demonstrate the generalizability of our results to balls of different weights and sizes, underscoring the robustness of our learning approach. This work, therefore, emphasizes the importance of the choice curriculum and challenges long-held notions about the need for tactile information to autonomously learn in-hand dexterous manipulation.