Orthogonal Deep Neural Networks
In this paper, we introduce the algorithms of Orthogonal Deep Neural Networks (OrthDNNs) to connect with recent interest of spectrally regularized deep learning methods. OrthDNNs are theoretically motivated by generalization analysis of modern DNNs, with the aim to find solution properties of networ...
Ausführliche Beschreibung
Bibliographische Detailangaben
Veröffentlicht in: | IEEE transactions on pattern analysis and machine intelligence. - 1979. - 43(2021), 4 vom: 01. Apr., Seite 1352-1368
|
1. Verfasser: |
Li, Shuai
(VerfasserIn) |
Weitere Verfasser: |
Jia, Kui,
Wen, Yuxin,
Liu, Tongliang,
Tao, Dacheng |
Format: | Online-Aufsatz
|
Sprache: | English |
Veröffentlicht: |
2021
|
Zugriff auf das übergeordnete Werk: | IEEE transactions on pattern analysis and machine intelligence
|
Schlagworte: | Journal Article
Research Support, Non-U.S. Gov't |