Spectral bias and task-model alignment explain generalization in kernel regression and infinitely wide neural networks

Canatar et al. propose a predictive theory of generalization in kernel regression applicable to real data. This theory explains various generalization phenomena observed in wide neural networks, which admit a kernel limit and generalize well despite being overparameterized.

Guardado en:
Detalles Bibliográficos
Autores principales: Abdulkadir Canatar, Blake Bordelon, Cengiz Pehlevan
Formato: article
Lenguaje:EN
Publicado: Nature Portfolio 2021
Materias:
Q
Acceso en línea:https://doaj.org/article/3fb570c6ce05419290b8cc1eebe16977
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
Descripción
Sumario:Canatar et al. propose a predictive theory of generalization in kernel regression applicable to real data. This theory explains various generalization phenomena observed in wide neural networks, which admit a kernel limit and generalize well despite being overparameterized.