Some Comparisons of Complexity in Dictionary-Based and Linear Computational Models

0350421 - ÚI 2011 RIV GB eng J - Journal Article
Gnecco, G. - Kůrková, Věra - Sanguineti, M.
Some Comparisons of Complexity in Dictionary-Based and Linear Computational Models.
Neural Networks. Roč. 24, č. 2 (2011), s. 171-182. ISSN 0893-6080. E-ISSN 1879-2782
R&D Projects: GA ČR GA201/08/1744
Grant - others:CNR - AV ČR project 2010-2012(XE) Complexity of Neural-Network and Kernel Computational Models
Institutional research plan: CEZ:AV0Z10300504
Keywords : linear approximation schemes * variable-basis approximation schemes * model complexity * worst-case errors * neural networks * kernel models
Subject RIV: IN - Informatics, Computer Science
Impact factor: 2.182, year: 2011

We compare upper bounds on worst-case errors in variable-basis approximation with lower bounds on such errors for any linear approximator. Using methods from nonlinear approximation and integral representations tailored to computational units, we describe some cases where neural networks outperform any linear approximator.
Permanent Link: http://hdl.handle.net/11104/0190434