A Comparative Analysis of Reinforcement Learning Methods
dc.date.accessioned | 2004-10-04T14:25:16Z | |
dc.date.accessioned | 2018-11-24T10:11:22Z | |
dc.date.available | 2004-10-04T14:25:16Z | |
dc.date.available | 2018-11-24T10:11:22Z | |
dc.date.issued | 1991-10-01 | en_US |
dc.identifier.uri | http://hdl.handle.net/1721.1/5978 | |
dc.identifier.uri | http://repository.aust.edu.ng/xmlui/handle/1721.1/5978 | |
dc.description.abstract | This paper analyzes the suitability of reinforcement learning (RL) for both programming and adapting situated agents. We discuss two RL algorithms: Q-learning and the Bucket Brigade. We introduce a special case of the Bucket Brigade, and analyze and compare its performance to Q in a number of experiments. Next we discuss the key problems of RL: time and space complexity, input generalization, sensitivity to parameter values, and selection of the reinforcement function. We address the tradeoffs between the built-in and learned knowledge and the number of training examples required by a learning algorithm. Finally, we suggest directions for future research. | en_US |
dc.format.extent | 13 p. | en_US |
dc.format.extent | 1444645 bytes | |
dc.format.extent | 1130480 bytes | |
dc.language.iso | en_US | |
dc.subject | reinforcement | en_US |
dc.subject | learning | en_US |
dc.subject | situated agents | en_US |
dc.subject | inputsgeneralization | en_US |
dc.subject | complexity | en_US |
dc.subject | built-in knowledge | en_US |
dc.title | A Comparative Analysis of Reinforcement Learning Methods | en_US |
Files in this item
Files | Size | Format | View |
---|---|---|---|
AIM-1322.pdf | 1.130Mb | application/pdf | View/ |
AIM-1322.ps | 1.444Mb | application/postscript | View/ |