Skip to Main content Skip to Navigation
Conference papers

Generating Natural Behaviors using Constructivist Algorithms

Olivier L. Georgeon 1, 2 Paul Robertson Jianyong Xue 1
1 SyCoSMA - Systèmes Cognitifs et Systèmes Multi-Agents
LIRIS - Laboratoire d'InfoRmatique en Image et Systèmes d'information
Abstract : We present a project to design interactive devices (smart displays, robots, etc.) capable of self-motivated learning through non-goal-directed interactive behaviors (e.g., curious, emotional, playful behaviors). We use and improve algorithms inspired by constructivist epistemology that we have designed previously. These algorithms incrementally learn sequential hierarchies of control loops in a bottom-up and open-ended fashion, and continuously reuse the learned higher-level control loops to generate increasingly complex behaviors that exhibit self-motivation. This project contributes to research in self-supervised learning because the learning is driven by low-level preferences that under-determine the device’s future behaviors, leaving room for individuation, which, in turn, opens the way to autonomy in learning.
Document type :
Conference papers
Complete list of metadata

https://hal.archives-ouvertes.fr/hal-02930211
Contributor : Olivier Georgeon <>
Submitted on : Friday, September 4, 2020 - 11:35:20 AM
Last modification on : Tuesday, June 1, 2021 - 2:08:10 PM

Identifiers

  • HAL Id : hal-02930211, version 1

Citation

Olivier L. Georgeon, Paul Robertson, Jianyong Xue. Generating Natural Behaviors using Constructivist Algorithms. International Workshop on Self-Supervised Learning, Feb 2020, Boston, United States. pp.5-14. ⟨hal-02930211⟩

Share

Metrics

Record views

46