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.
Document type :
Conference papers
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-01931713
Contributor : David Filliat <>
Submitted on : Thursday, November 29, 2018 - 9:56:25 AM
Last modification on : Wednesday, July 3, 2019 - 10:48:05 AM

File

S_RL_Toolbox.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01931713, version 1

Citation

Antonin Raffin, Ashley Hill, René Traoré, Timothée Lesort, Natalia Díaz-Rodríguez, et al.. S-RL Toolbox: Environments, Datasets and Evaluation Metrics for State Representation Learning. NIPS 2018 Deep RL workshop, Dec 2018, Montreal, Canada. ⟨hal-01931713⟩

Share

Metrics

Record views

40

Files downloads

118