S-RL Toolbox: Environments, Datasets and Evaluation Metrics for State Representation Learning

Abstract

State representation learning aims at learning compact representations from raw observations in robotics and control applications. Approaches used for this objective are auto-encoders, learning forward models, inverse dynamics or learning using generic priors on the state characteristics. However, the diversity in applications and methods makes the field lack standard evaluation datasets, metrics and tasks. This paper provides a set of environments, data generators, robotic control tasks, metrics and tools to facilitate iterative state representation learning and evaluation in reinforcement learning settings.

arXiv commentGithub repo: https://github.com/araffin/robotics-rl-srl Documentation: https://s-rl-toolbox.readthedocs.io/en/latest/

References in corpus (5)

• Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor
• Multi-Goal Reinforcement Learning: Challenging Robotics Environments and Request for Research
• Simple random search provides a competitive approach to reinforcement learning
• Loss is its own Reward: Self-Supervision for Reinforcement Learning
• Decoupling Dynamics and Reward for Transfer Learning

Cited by in corpus (11)

• Acquisition of Chess Knowledge in AlphaZero
• Mid-Level Visual Representations Improve Generalization and Sample Efficiency for Learning Visuomotor Policies
• DisCoRL: Continual Reinforcement Learning via Policy Distillation
• Continual State Representation Learning for Reinforcement Learning using Generative Replay
• Robust Policies via Mid-Level Visual Representations: An Experimental Study in Manipulation and Navigation
• Learning to Navigate Using Mid-Level Visual Priors
• DREAM Architecture: a Developmental Approach to Open-Ended Learning in Robotics
• Continual Learning: Tackling Catastrophic Forgetting in Deep Neural Networks with Replay Processes
• Analytic Manifold Learning: Unifying and Evaluating Representations for Continuous Control
• State Representation Learning from Demonstration
• Supervise Thyself: Examining Self-Supervised Representations in Interactive Environments