Please use this identifier to cite or link to this item:
http://hdl.handle.net/10261/30062
Share/Export:
![]() |
|
Visualizar otros formatos: MARC | Dublin Core | RDF | ORE | MODS | METS | DIDL | DATACITE | |
Title: | Probability density estimation of the Q function for reinforcement learning |
Authors: | Agostini, Alejandro CSIC ORCID; Celaya, Enric CSIC | Keywords: | Gaussian mixture models On-line EM Reinforcement learning Continuous state-action spaces Intelligent robots and autonomous agents Machine learning |
Issue Date: | 2009 | Citation: | Technical Report IRI-TR-09-06, Institut de Robòtica i Informàtica Industrial, CSIC-UPC, 2009. | Abstract: | Performing Q-Learning in continuous state-action spaces is a problem still unsolved for many complex applications. The Q function may be rather complex and can not be expected to fit into a predefined parametric model. In addition, the function approximation must be able to cope with the high non-stationarity of the estimated q values, the on-line nature of the learning with a strongly biased sampling to convergence regions, and the large amount of generalization required for a feasible implementation. To cope with these problems local, non-parametric function approximations seem more suitable than global parametric ones. A kind of function approximation that is gaining special interest in the field of machine learning are those based on densities. Estimating densities provides more information than simple function approximations which can be used to deal with the Reinforcement Learning problems. For instance, density estimation permits to know the actual distribution of the q values for any given state-action, and provides information about how many data has been collected in different regions of the domain. In this work we propose a Q-Learning approach for continuous state-action spaces based on joint density estimations. The density distribution is represented with a Gaussian Mixture Model using an on-line version of the Expectation-Maximization algorithm. We propose a method that handles the biased sampling problem with good performance. Experiments performed on a test problem show remarkable improvements over previous published results. | URI: | http://hdl.handle.net/10261/30062 |
Appears in Collections: | (IRII) Informes y documentos de trabajo |
Show full item record
Review this work
Page view(s)
336
checked on May 24, 2022
Download(s)
643
checked on May 24, 2022
Google ScholarTM
Check
WARNING: Items in Digital.CSIC are protected by copyright, with all rights reserved, unless otherwise indicated.