visitante ::
identificación
|
|||||||||||||||
Buscar | Enviar | Ayuda | Servicio de Bibliotecas | Sobre el DDD | Català English Español |
Página principal > Libros y colecciones > Capítulos de libros > Rotate your networks : |
Publicación: | Institute of Electrical and Electronics Engineers (IEEE), cop.2018 |
Descripción: | 7 pàg. |
Resumen: | In this paper we propose an approach to avoiding catastrophic forgetting in sequential task learning scenarios. Our technique is based on a network reparameterization that approximately diagonalizes the Fisher Information Matrix of the network parameters. This reparameterization takes the form of a factorized rotation of parameter space which, when used in conjunction with Elastic Weight Consolidation (which assumes a diagonal Fisher Information Matrix), leads to significantly better performance on lifelong learning of sequential tasks. Experimental results on the MNIST, CIFAR-100, CUB-200 and Stanford-40 datasets demonstrate that we significantly improve the results of standard elastic weight consolidation, and that we obtain competitive results when compared to the state-of-the-art in lifelong learning without forgetting. |
Ayudas: | European Commission 6655919 European Commission 777720 Agencia Estatal de Investigación TIN2017-88709-R Agencia Estatal de Investigación TIN2016-79717-R Ministerio de Economía y Competitividad PCIN-2015-251 Agència de Gestió d'Ajuts Universitaris i de Recerca 2018/FI B1-00198 |
Derechos: | Tots els drets reservats. |
Lengua: | Anglès |
Documento: | Capítol de llibre ; recerca ; Versió acceptada per publicar |
Materia: | Task analysis ; Training ; Training data ; Neural networks ; Data models ; Computer vision ; Standards |
Publicado en: | 2018 24th International Conference on Pattern Recognition (ICPR), 2018, p. 2262-2268, ISBN 978-1-5386-3788-3 |
Postprint 9 p, 611.7 KB |