Mason Archival Repository Service

Adaptive Anchoring Discretization for Learnable Evolution Model: The ANCHOR Method

Show simple item record

dc.contributor.author Michalski, Ryszard S.
dc.contributor.author Cervone, Guido
dc.date.accessioned 2006-11-03T18:17:12Z
dc.date.available 2006-11-03T18:17:12Z
dc.date.issued 2001-05 en_US
dc.identifier.citation Michalski, R. S. and Cervone, G., "Adaptive Anchoring Discretization for Learnable Evolution Model: The ANCHOR Method," Reports of the Machine Learning and Inference Laboratory, MLI 01-3, George Mason University, Fairfax, VA, 2001. en_US
dc.identifier.uri https://hdl.handle.net/1920/1473
dc.description.abstract To apply a symbolic learning method to learning in a continuous representation space, the variables spanning the space need to be discretized. When the space is very large, a problem arises as to how to determine a discretization scheme for each variable that is both efficient and effective. This task is particularly important when applying Learnable Evolution Model to optimization problems with very large number of continuous variables. The presented method, called ANCHOR, starts with a low discretization precision of the variables, and then increases the discretization precision in the subranges indicated by the analysis of the descriptions learned using a lower precision. The method has been incorporated in the LEM2 system implementing the Learnable Evolution Model. Experiments with ANCHOR have demonstrated a significant advantage of the method over a fixed discretization method, and enabled LEM2 to optimize functions of large number of continuous variables very effectively.
dc.description.sponsorship This research has been conducted in the Machine Learning and Inference Laboratory at the School of Computational Sciences, George Mason University. The Laboratory's research has been supported in part by the National Science Foundation under Grants No. IIS-9906858 and IIS-0097476, and in part by the UMBC/LUCITE #32 grant. en_US
dc.format.extent 2406 bytes
dc.format.extent 455581 bytes
dc.format.extent 89795 bytes
dc.format.mimetype text/xml
dc.format.mimetype application/postscript
dc.format.mimetype application/pdf
dc.language.iso en_US en_US
dc.relation.ispartofseries P 01-2 en_US
dc.relation.ispartofseries MLI 01-3 en_US
dc.title Adaptive Anchoring Discretization for Learnable Evolution Model: The ANCHOR Method en_US
dc.type Technical report en_US


Files in this item

This item appears in the following Collection(s)

Show simple item record

Search MARS


Browse

My Account

Statistics