Home // COGNITIVE 2011, The Third International Conference on Advanced Cognitive Technologies and Applications // View article

Improvements on Relational Reinforcement Learning to Solve Joint Attention

Authors:
Renato Silva
Roseli Romero

Keywords: joint attention; relational reinforcement learning; social robots; shared attention.

Abstract:
The joint attention is an important cognitive function that human beings learn in childhood. This nonverbal communication is very important for a person understands other individuals and the environment during the interaction. Because of this, it is essential that the robots learn this skill to be inserted in the environment and interact socially. In this article, we have enhanced a robotic architecture, which is inspired on Behavior Analysis, to provide the capacity of learning joint attention on robots or agents using only relational reinforcement learning when the environment changes. Then,a set of empirical evaluations has been conducted in the social interactive simulator for performing the task of joint attention. The performance of this algorithm have been compared with the Q-Learning algorithm, contingency learning algorithm and ETG algorithm. The experimental results show that this algorithm solves the problems of learning and makes the architecture with greater flexibility to insert new modules.

Pages: 63 to 69

Copyright: Copyright (c) IARIA, 2011

Publication date: September 25, 2011

Published in: conference

ISSN: 2308-4197

ISBN: 978-1-61208-155-7

Location: Rome, Italy

Dates: from September 25, 2011 to September 30, 2011