0404821 - UIVT-O 20030183 RIV AT eng C - Conference Paper (international conference)
Kůrková, Věra - Sanguineti, M.
Neural Network Learning as Approximate Optimization.
Artificial Neural Nets and Genetic Algorithms. Wien: SpringerVerlag, 2003 - (Pearson, D.; Steele, N.; Albrecht, R.), s. 53-57. ISBN 3-211-00743-1.
[ICANNGA'2003 /6./. Roanne (FR), 23.04.2003-25.04.2003]
R&D Projects: GA ČR GA201/02/0428
Grant - others:IT-CZ Area MC6(XX) Project 22
Institutional research plan: AV0Z1030915
Keywords : neural networks * learning from data * approximate optimization
Subject RIV: BA - General Mathematics
DOI: https://doi.org/10.1007/978-3-7091-0646-4_11

Learning from data will be studied in the framework of approximate minimization of regularized empirical error functionals. There will be derived estimates of speed of convergence of infima achievable over approximations of an admissible set to a global infimum. The results will be applied to empirical error functionals regularized using stabilizers defined as squares of norms in reproducing kernel Hilbert spaces.

Permanent Link: http://hdl.handle.net/11104/0125054