%0 Journal Article
%A YANG Juan
%A ZHU Fan
%T Fast TMRM: efficient multi-task recommendation model
%D 2020
%R 10.19682/j.cnki.1005-8885.2020.0025
%J Journal of China Universities of Posts and Telecommunications
%P 13-22
%V 27
%N 5
%X <span>An improved multi-task learning recommendation algorithm—fast two-stage multi-task recommendation model&nbsp;</span><span>boosted feature selection (Fast TMRM) is proposed based on auto-encoders in this paper. Compared to previous&nbsp;</span><span>work, Fast TMRM improves the convergence speed and accuracy of training. In addition, Fast TMRM builds on&nbsp;</span><span>previous work to introduce the auto-encoder to encode the important feature combination vector. That is how it can&nbsp;</span><span>be used for the training of multi-task learning, which helps to improve the training efficiency of the model by nearly&nbsp;</span><span>67%. Finally, the nearest neighbor search is used to restore important feature expression.</span>
%U https://jcupt.bupt.edu.cn/EN/10.19682/j.cnki.1005-8885.2020.0025