Parallel Coordinate Descent Newton Method for Efficient [Formula Omitted]-Regularized Loss Minimization

The recent years have witnessed advances in parallel algorithms for large-scale optimization problems. Notwithstanding the demonstrated success, existing algorithms that parallelize over features are usually limited by divergence issues under high parallelism or require data preprocessing to allevia...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	IEEE transaction on neural networks and learning systems 2019-01, Vol.30 (11), p.3233
Hauptverfasser:	Yatao An Bian, Li, Xiong, Liu, Yuncai, Ming-Hsuan Yang
Format:	Artikel
Sprache:	eng
Schlagworte:	Algorithms Convergence Data transfer (computers) Descent Divergence Iterative methods Newton methods Optimization Parallel processing Preprocessing Searching Support vector machines Synchronism Synchronization
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!