Field-wise Embedding Size Search via Structural Hard Auxiliary Mask Pruning for Click-Through Rate Prediction

Feature embeddings are one of the most essential steps when training deep learning based Click-Through Rate prediction models, which map high-dimensional sparse features to dense embedding vectors. Classic human-crafted embedding size selection methods are shown to be "sub-optimal" in term...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	arXiv.org 2022-08
Hauptverfasser:	Tesi Xiao, Xia, Xiao, Chen, Ming, Chen, Youlong
Format:	Artikel
Sprache:	eng
Schlagworte:	Columns (structural) Computing costs Embedding Prediction models Pruning Searching
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!