BriefGPT.xyz
Jun, 2017
基于Top-k Gains的循环神经网络会话推荐
Recurrent Neural Networks with Top-k Gains for Session-based Recommendations
HTML
PDF
Balázs Hidasi, Alexandros Karatzoglou
TL;DR
该研究介绍了针对会话推荐的 RNN 模型的新型排名损失函数，与其他方法相比，这些损失函数表现出更好的性能，并且可以通过进一步细化和改进，实现相对于先前的 RNN 解决方案提高了 35% 的 MRR 和 Recall @20，比经典的协作过滤方法提高了 53% 的性能。
Abstract
rnns
have been shown to be excellent models for sequential data and in particular for session-based user behavior. The use of
rnns
provides impressive performance benefits over classical methods in
→