Difference between revisions of "2016 DeepNeuralNetworksforYoutubeRec"

From GM-RKB
Jump to: navigation, search
m (Gmelli moved page Test:2016 DeepNeuralNetworksforYoutubeRec to 2016 DeepNeuralNetworksforYoutubeRec without leaving a redirect)
Line 1: Line 1:
 
* ([[2016_DeepNeuralNetworksforYoutubeRec|Covington et al., 2016]]) ⇒ [[author::Paul Covington]], [[author::Jay Adams]], and [[author::Emre Sargin]]. ([[year::2016]]). “[https://research.google.com/pubs/archive/45530.pdf Deep Neural Networks for Youtube Recommendations].” In: Proceedings of the 10th ACM conference on recommender systems.  
 
* ([[2016_DeepNeuralNetworksforYoutubeRec|Covington et al., 2016]]) ⇒ [[author::Paul Covington]], [[author::Jay Adams]], and [[author::Emre Sargin]]. ([[year::2016]]). “[https://research.google.com/pubs/archive/45530.pdf Deep Neural Networks for Youtube Recommendations].” In: Proceedings of the 10th ACM conference on recommender systems.  
  
<B>Subject Headings:</B>  
+
<B>Subject Headings:</B> [[Deep Candidate Generation Model]], [[Deep Item Ranking Model]].
  
 
==Notes==
 
==Notes==
Line 14: Line 14:
 
===Abstract===
 
===Abstract===
  
[[YouTube]] represents one of the [[largest scale]] and most sophisticated [[industrial recommendation system]]s in existence. </s>
+
[[YouTube]] represents one of the [[large scale recommender system|largest scale]] and most sophisticated [[industrial recommendation system]]s in existence. </s>
[[In this paper, we]] describe [[the system]] at a [[high level]] and focus on the dramatic [[performance]] improvements brought by [[deep learning]]. </s>
+
[[In this paper, we]] describe [[the system]] at a [[high level]] and focus on the dramatic [[performance improvement]]s brought by [[deep learning recommendation algorithm|deep learning]]. </s>
[[The paper is split]] according to the [[classic]] [[two-stage information retrieval dichotomy]]: first, [[we]] detail a [[deep candidate generation model]] and then describe a separate [[deep ranking model]]. </s>
+
[[The paper]] is split according to the [[classic]] [[two-stage]] [[information retrieval algorithm|information retrieval dichotomy]]: first, [[we]] detail a [[deep candidate generation model]] and then describe a separate [[deep ranking model]]. </s>
[[We]] also provide practical [[lesson]]s and insights derived from designing, [[iterating]] and [[maintaining a massive recommendation system]] with enormous [[user-facing impact]]. </s>
+
[[We]] also provide practical [[lesson]]s and insights derived from [[designing]], [[iterating]] and [[maintaining]] a [[massive recommendation system]] with enormous [[user-facing impact]]. </s>
  
 
==References==
 
==References==

Revision as of 22:30, 26 March 2020

Subject Headings: Deep Candidate Generation Model, Deep Item Ranking Model.

Notes

Cited By


Quotes

Abstract

YouTube represents one of the largest scale and most sophisticated industrial recommendation systems in existence. In this paper, we describe the system at a high level and focus on the dramatic performance improvements brought by deep learning. The paper is split according to the classic two-stage information retrieval dichotomy: first, we detail a deep candidate generation model and then describe a separate deep ranking model. We also provide practical lessons and insights derived from designing, iterating and maintaining a massive recommendation system with enormous user-facing impact.

References

;

 AuthorvolumeDate ValuetitletypejournaltitleUrldoinoteyear
2016 DeepNeuralNetworksforYoutubeRecPaul Covington
Jay Adams
Emre Sargin
Deep Neural Networks for Youtube Recommendations2016
AuthorPaul Covington +, Jay Adams + and Emre Sargin +
titleDeep Neural Networks for Youtube Recommendations +
year2016 +