Resource title

Maximal width learning of binary functions

Resource image

image for OpenScout resource :: Maximal width learning of binary functions

Resource description

This paper concerns learning binary-valued functions defined on IR, and investigates how a particular type of ‘regularity’ of hypotheses can be used to obtain better generalization error bounds. We derive error bounds that depend on the sample width (a notion similar to that of sample margin for real-valued functions). This motivates learning algorithms that seek to maximize sample width.

Resource author

Resource publisher

Resource publish date

Resource language

en

Resource content type

Resource resource URL

http://eprints.lse.ac.uk/13809/

Resource license