Deep unsupervised state representation learning with robotic priors: a robustness analysis - ENSTA Paris - École nationale supérieure de techniques avancées Paris Accéder directement au contenu
Communication Dans Un Congrès Année : 2019

Deep unsupervised state representation learning with robotic priors: a robustness analysis

Résumé

Our understanding of the world depends highly on our capacity to produce intuitive and simplified representations which can be easily used to solve problems. We reproduce this simplification process using a neural network to build a low dimensional state representation of the world from images acquired by a robot. As in Jonschkowski et al. 2015, we learn in an unsupervised way using prior knowledge about the world as loss functions called robotic priors and extend this approach to high dimension richer images to learn a 3D representation of the hand position of a robot from RGB images. We propose a quantitative evaluation metric of the learned representation that uses nearest neighbors in the state space and allows to assess its quality and show both the potential and limitations of robotic priors in realistic environments. We augment image size, add distractors and domain randomization, all crucial components to achieve transfer learning to real robots. Finally, we also contribute a new prior to improve the robustness of the representation. The applications of such low dimensional state representation range from easing reinforcement learning (RL) and knowledge transfer across tasks, to facilitating learning from raw data with more efficient and compact high level representations. The results show that the robotic prior approach is able to extract high level representation as the 3D position of an arm and organize it into a compact and coherent space of states in a challenging dataset.
Fichier principal
Vignette du fichier
2019 IJCNN Budapest Unsupervised state representation learning with robotic priors- a robustness analysis.pdf (5.44 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02381375 , version 1 (29-11-2019)

Identifiants

Citer

Timothée Lesort, Mathieu Seurin, Xinrui Li, Natalia Díaz-Rodríguez, David Filliat. Deep unsupervised state representation learning with robotic priors: a robustness analysis. IJCNN 2019 - International Joint Conference on Neural Networks, Jul 2019, Budapest, Hungary. pp.1-8, ⟨10.1109/IJCNN.2019.8852042⟩. ⟨hal-02381375⟩
83 Consultations
159 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More