PME: pruning-based multi-size embedding for recommender systems

Embedding is widely used in recommendation models to learn feature representations. However, the traditional embedding technique that assigns a fixed size to all categorical features may be suboptimal due to the following reasons. In recommendation domain, the majority of categorical features' embed...

Popoln opis

Bibliografske podrobnosti

Principais autores:	Zirui Liu, Qingquan Song, Li Li, Soo-Hyun Choi, Rui Chen, Xia Hu
Format:	Artigo
Jezik:	Inglês
Izdano:	Frontiers Media S.A. 2023-06-01
Serija:	Frontiers in Big Data
Teme:	neural network recommender system embedding compression pruning scalability
Online dostop:	https://www.frontiersin.org/articles/10.3389/fdata.2023.1195742/full
Oznake:	Označite Brez oznak, prvi označite!