Abstract
In this paper we present an approach that combines an artificial agent constructed from a general solution to the reinforcement learning problem and the Canadian Fire Weather Index. The study illustrates that the integration of AIXI and the problem of action selection in wildfire management scenarios simultaneously downscales the AIXI model and offers a practical demonstration of applied AIXI theory, while also illustrating the potential usefulness of such a tool in fire and natural resource management.
| Original language | English |
|---|---|
| Title of host publication | International Conference on Intelligent Agents, Web Technologies and Internet Commerce - IAWTIC'2014 Jointly with International Conference on Computational for Modelling, Control and Automation - CIMCA'2014 |
| Place of Publication | Online |
| Publisher | Conference |
| Pages | 31-37 |
| Number of pages | 7 |
| ISBN (Print) | 9780646935942 |
| Publication status | Published - 2014 |
| Event | International Conference on Intelligent Agents, Web Technologies and Internet Commerce - IAWTIC'2014 Jointly with International Conference on Computational for Modelling, Control and Automation - CIMCA'2014 - Online Conference, Vienna, Austria Duration: 10 Dec 2014 → 12 Dec 2014 http://www.cimca-conf.org/iawtic14/ |
Conference
| Conference | International Conference on Intelligent Agents, Web Technologies and Internet Commerce - IAWTIC'2014 Jointly with International Conference on Computational for Modelling, Control and Automation - CIMCA'2014 |
|---|---|
| Country | Austria |
| City | Vienna |
| Period | 10/12/14 → 12/12/14 |
| Internet address |
Fingerprint
Cite this
Omari, R., Newth, D., Böhm, M., & MOHAMMADIAN, M. (2014). An Application of a Monte-Carlo AIXI Approximation in Ecological Fire Management. In International Conference on Intelligent Agents, Web Technologies and Internet Commerce - IAWTIC'2014 Jointly with International Conference on Computational for Modelling, Control and Automation - CIMCA'2014 (pp. 31-37). Online: Conference.