Title :
Performance Evaluation of TD-Learning Methods for Bandwidth Provisioning
Author :
Jahanshahi, M. ; Meybodi, M.R.
Author_Institution :
Islamic Azad Univ., Tehran
Abstract :
Q-learning and SARSA are two methods of TD- learning. Researchers interested in this field proposed the Eligibility concept in order to speed up Q-learning and SARSA. They proved their claim by running the algorithms in a static environment. Authors of this paper have used Q-learning, SARSA and also their eligibility versions for bandwidth provisioning in DiffServ networks that is an absolutely dynamic environment. Performance of these methods in this absolutely dynamic environment is evaluated.
Keywords :
DiffServ networks; performance evaluation; DiffServ network; Q-learning; bandwidth provisioning; dynamic environment evaluation; eligibility concept; performance evaluation; temporal-difference learning; Artificial intelligence; Bandwidth; Computer architecture; Delay; Diffserv networks; Information technology; Learning systems; Stochastic processes; Testing; Videoconference;
Conference_Titel :
Computer and Information Technology, 2007. CIT 2007. 7th IEEE International Conference on
Conference_Location :
Aizu-Wakamatsu, Fukushima
Print_ISBN :
978-0-7695-2983-7
DOI :
10.1109/CIT.2007.131
