2011/08/01 by Sao Mai Nguyen, Adrien Baranes, Pierre-Yves Oudeyer
Computer Science · #Active learning (machine learning) #Bootstrapping (finance) #Coupling (piping) #Hybrid learning #Intrinsic motivation #Machine Learning and Algorithms #Mobile Crowdsensing and Crowdsourcing #Reinforcement Learning in Robotics #Repertoire #Social learning #cs.AI #cs.LG #cs.RO
paper · pdf · doi:10.1109/devlrn.2011.6037329
published as 2011 IEEE International Conference on Development and Learning (ICDL) · IEEE International Conference on Development and Learning, Frankfurt : Germany (2011)
openalex publication_date 2011/08/01 · arxiv created 2011/12/08 · openalex created_date 2016/06/24 · arxiv updated 2019/02/01 · openalex updated_date 2026/08/05
This paper studies the coupling of internally guided learning and social interaction, and more specifically the improvement owing to demonstrations of the learning by intrinsic motivation. We present Socially Guided Intrinsic Motivation by Demonstration (SGIM-D), an algorithm for learning in continuous, unbounded and non-preset environments. After introducing social learning and intrinsic motivation, we describe the design of our algorithm, before showing through a fishing experiment that SGIM-D efficiently combines the advantages of social learning and intrinsic motivation to gain a wide repertoire while being specialised in specific subspaces.