dc.contributor.author Barnard, Etienne en_US dc.date.accessioned 2014-08-15T12:56:30Z dc.date.available 2014-08-15T12:56:30Z dc.date.issued 2011 en_US dc.identifier.citation Barnard, E. 2011. Determination and the no-free-lunch paradox. Neural Computation, 23(7):1899-1909. [https://doi.org/10.1162/NECO_a_00137] en_US dc.identifier.issn 0899-7667 dc.identifier.issn 1530-888X dc.identifier.uri http://hdl.handle.net/10394/11187 dc.identifier.uri https://doi.org/10.1162/NECO_a_00137 dc.description.abstract We discuss the no-free-lunch NFL theorem for supervised learning as a logical paradox—that is, as a counterintuitive result that is correctly proven from apparently incontestable assumptions. We show that the uniform prior that is used in the proof of the theorem has a number of unpalatable consequences besides the NFL theorem, and propose a simple definition of determination (by a learning set of given size) that casts additional suspicion on the utility of this assumption for the prior. Whereas others have suggested that the assumptions of the NFL theorem are not practically realistic, we show these assumptions to be at odds with supervised learning in principle. This analysis suggests a route toward the establishment of a more realistic prior probability for use in the extended Bayesian framework. dc.language en dc.publisher MIT Press dc.title Determination and the no-free-lunch paradox en_US dc.type Article dc.contributor.researchID 21021287 - Barnard, Etienne
﻿

## Files in this item

FilesSizeFormatView

There are no files associated with this item.

Theme by