On the Use of Intrinsic Motivation for Visual Saliency Learning

Céline Craye 1, 2, 3 David Filliat 2, 3 Jean-François Goudou 1
2 Flowers - Flowing Epigenetic Robots and Systems
Inria Bordeaux - Sud-Ouest, U2IS - Unité d'Informatique et d'Ingénierie des Systèmes
Abstract : The use of intrinsic motivation for the task of learning sensori-motor properties has received a lot of attention over the last few years, but only little work has been provided toward using intrinsic motivation for the task of learning visual signals. In this paper, we propose to apply the main ideas of the Intelligent Adaptive Curiosity (IAC) for the task of visual saliency learning. We here present RL-IAC, an adapted version of IAC that uses reinforcement learning to deal with time consuming displacements while actively learning saliency based on local learning progress. We also introduce the use of a backward evaluation to deal with a learner that is shared between several regions. We demonstrate the good performance of RL-IAC compared to other exploration techniques, and we discuss the performance of other intrinsic motivation sources instead of learning progress in our problem.
Document type :
Conference papers
Complete list of metadatas

Cited literature [23 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01370850
Contributor : David Filliat <>
Submitted on : Friday, September 23, 2016 - 1:42:33 PM
Last modification on : Wednesday, July 3, 2019 - 10:48:05 AM

File

Craye_ICDL16_final.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01370850, version 1

Citation

Céline Craye, David Filliat, Jean-François Goudou. On the Use of Intrinsic Motivation for Visual Saliency Learning. Sixth Joint IEEE International Conference Developmental Learning and Epigenetic Robotics (ICDL-EPIROB), Sep 2016, Cergy-Pontoise, France. ⟨hal-01370850⟩

Share

Metrics

Record views

296

Files downloads

275