Antonin Raffin antonin.raffin@ensta-paristech.fr Ashley Hill ashley.hill@ensta-paristech.fr René Traoré rene.traore@ensta-paristech.fr Timothée Lesort timothee.lesort@ensta-paristech.fr Natalia Díaz-Rodríguez natalia.diaz@ensta-paristech.fr David Filliat david.filliat@ensta-paristech.fr U2IS, ENSTA ParisTech / INRIA FLOWERS Team http: // flowers. inria. fr
State representation learning aims at learning compact representations from raw observations in robotics and control applications.
Approaches used for this objective are autoencoders, 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.
Keywords: Deep learning, reinforcement learning, state representation learning, robotic priors