Shapechanger: Environments for Transfer Learning
S\'ebastien M. R. Arnold, Tsam Kiu Pun, Th\'eo-Tim J. Denisart and, Francisco J. Valero-Cuevas
TL;DR
Shapechanger is an open-source library that facilitates transfer reinforcement learning across various robotic environments, enabling knowledge transfer from simulation to simulation, simulation to real, and real to real for continuous tasks.
Contribution
It introduces a flexible library supporting diverse transfer learning scenarios in robotics, addressing a gap in existing tools.
Findings
Supports transfer from simulation to simulation, simulation to real, and real to real.
Handles tasks with continuous states and actions.
Open-source and actively developed.
Abstract
We present Shapechanger, a library for transfer reinforcement learning specifically designed for robotic tasks. We consider three types of knowledge transfer---from simulation to simulation, from simulation to real, and from real to real---and a wide range of tasks with continuous states and actions. Shapechanger is under active development and open-sourced at: https://github.com/seba-1511/shapechanger/.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsReinforcement Learning in Robotics · Robot Manipulation and Learning · Adversarial Robustness in Machine Learning