Decoupling feature extraction from policy learning: assessing benefits of state representation learning in goal based robotics

Abstract

Scaling end-to-end reinforcement learning to control real robots from vision presents a series of challenges, in particular in terms of sample efficiency. Against end-to-end learning, state representation learning can help learn a compact, efficient and relevant representation of states that speeds up policy learning, reducing the number of samples needed, and that is easier to interpret. We evaluate several state representation learning methods on goal based robotics tasks and propose a new unsupervised model that stacks representations and combines strengths of several of these approaches. This method encodes all the relevant features, performs on par or better than end-to-end learning with better sample efficiency, and is robust to hyper-parameters change.

arXiv commentGithub repo: https://github.com/araffin/srl-zoo Documentation: https://srl-zoo.readthedocs.io/en/latest/, As part of SRL-Toolbox: https://s-rl-toolbox.readthedocs.io/en/latest/. Accepted to the Workshop on Structure & Priors in Reinforcement Learning at ICLR 2019

Cited by in corpus (7)

• Acquisition of Chess Knowledge in AlphaZero
• A survey on intrinsic motivation in reinforcement learning
• ShieldNN: A Provably Safe NN Filter for Unsafe NN Controllers
• Learning Group Structure and Disentangled Representations of Dynamical Environments
• Learning from learning machines: a new generation of AI technology to meet the needs of science
• Scope and Sense of Explainability for AI-Systems
• POAR: Efficient Policy Optimization via Online Abstract State Representation Learning