Anthony, Martin and Ratsaby, Joel (2010) Maximal width learning of binary functions. Theoretical Computer Science, 411 (1). pp. 138-147. ISSN 0304-3975
This paper concerns learning binary-valued functions defined on, 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 analogous to that of sample margin for real-valued functions). This motivates learning algorithms that seek to maximize sample width.
|Additional Information:||© 2009 Elsevier|
|Library of Congress subject classification:||Q Science > QA Mathematics|
|Sets:||Departments > Mathematics|
|Date Deposited:||11 Aug 2010 11:15|
Actions (login required)
|Record administration - authorised staff only|