2 Hits in 4.2 sec

Grounding Language to Autonomously-Acquired Skills via Goal Generation [article]

Ahmed Akakzia, Cédric Colas, Pierre-Yves Oudeyer, Mohamed Chetouani, Olivier Sigaud
2021 arXiv   pre-print
DECSTR is an intrinsically motivated learning agent endowed with an innate semantic representation describing spatial relations between physical objects.  ...  Besides, their direct language condition cannot account for the goal-directed behavior of pre-verbal infants and strongly limits the expression of behavioral diversity for a given language input.  ...  Cédric Colas is partly funded by the French Ministère des Armées -Direction Générale de l'Armement.  ... 
arXiv:2006.07185v3 fatcat:3p653fah5zbxpghexuumq2nwua

Towards Teachable Autotelic Agents [article]

Olivier Sigaud and Ahmed Akakzia and Hugo Caselles-Dupré and Cédric Colas and Pierre-Yves Oudeyer and Mohamed Chetouani
2022 arXiv   pre-print
Autonomous discovery and direct instruction are two distinct sources of learning in children but education sciences demonstrate that mixed approaches such as assisted discovery or guided play result in  ...  The checklist allows us to precisely pinpoint the various limitations of current reinforcement learning agents and to identify the promising first steps towards TAA.  ...  The decstr agent solves the latter issue by introducing an abstract goal representation layer in the architecture where a goal is expressed as general predicates.  ... 
arXiv:2105.11977v2 fatcat:w37mjxeaafefnko3a64on7l3uu