Multi-task model training method, promotion content processing method and related device

The invention provides a multi-task model training method, a promotion content processing method and a related device.A multi-task model comprises an auxiliary network, a core network and a look-forward gradient network, the method comprises the steps that sample feature vectors are obtained, and th...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: ZHAO XIUYING, WU YINCHU, SHE QI
Format: Patent
Sprache:chi ; eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The invention provides a multi-task model training method, a promotion content processing method and a related device.A multi-task model comprises an auxiliary network, a core network and a look-forward gradient network, the method comprises the steps that sample feature vectors are obtained, and the sample feature vectors are used for representing behaviors of a user on promotion content; inputting the sample feature vector into an auxiliary exclusive network to obtain an auxiliary feature vector; inputting the auxiliary feature vector into a look-ahead gradient network to obtain a gradient feature vector, wherein the gradient feature vector represents the contribution degree of the auxiliary feature vector to the optimization target of the core network; fusing the gradient feature vector and the auxiliary feature vector, and inputting into an auxiliary output network to obtain the output of the auxiliary output network; and updating the weight of the auxiliary network according to the output of the auxiliar