RT Journal Article
T1 Learning to Avoid Risky Actions
A1 Malfaz Vázquez, María Ángeles
A1 Salichs Sánchez-Caballero, Miguel
AB When a reinforcement learning agent executes actions that can cause frequent damage to itself, it can learn, by using Q-learning, that these actions must not be executed again. However, there are other actions that do not cause damage frequently but only once in a while, for example, risky actions such as parachuting. These actions may imply punishment to the agent and, depending on its personality, it would be better to avoid them. Nevertheless, using the standard Q-learning algorithm, the agent is not able to learn to avoid them, because the result of these actions can be positive on average. In this article, an additional mechanism of Q-learning, inspired by the emotion of fear, is introduced in order to deal with those risky actions by considering the worst results. Moreover, there is a daring factor for adjusting the consideration of the risk. This mechanism is implemented on an autonomous agent living in a virtual environment. The results present the performance of the agent with different daring degrees.
PB Taylor & Francis Group
SN 0196-9722 (print)
SN 1087-6553 (online)
YR 2011
FD 2011-12
LK https://hdl.handle.net/10016/18621
UL https://hdl.handle.net/10016/18621
LA eng
NO The funds provided by the Spanish Governmentthrough the project called “A New Approach to Social Robotics” (AROS), ofMICINN (Ministry of Science and Innovation) and through the RoboCity2030-IICMproject (S2009/DPI-1559), funded by Programas de Actividades I+D en laComunidad de Madrid and cofunded by Structural Funds of the EU.
DS e-Archivo
RD 29 jun. 2024