Learning States Representations in POMDP

Abstract : We propose to deal with sequential processes where only partial observations are available by learning a latent representation space on which policies may be accurately learned.
Document type :
Poster communications
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01222594
Contributor : Lip6 Publications <>
Submitted on : Friday, October 30, 2015 - 11:31:59 AM
Last modification on : Thursday, March 21, 2019 - 1:13:24 PM

Links full text

Identifiers

  • HAL Id : hal-01222594, version 1
  • ARXIV : 1312.6042

Citation

Gabriella Contardo, Ludovic Denoyer, Thierry Artières, Patrick Gallinari. Learning States Representations in POMDP. International Conference on Learning Representations, ICLR 2014, Apr 2014, Banff, Canada. 2014. ⟨hal-01222594⟩

Share

Metrics

Record views

91