重磅|推荐系统顶会RecSys2018最佳论文奖出炉!因果嵌入推荐与用户研究成为焦点

【导读】第12届推荐系统顶级会议ACM RecSys在10月2日到7日的加拿大温哥华举行最近官网大会公布了最佳长短文,包括来自Criteo AI Labs的因果嵌入推荐与德国Duisburg-Essen大学的用户评估中物品消费的影响

官网地址:

https://recsys.acm.org/best-papers/



最佳长论文

RecSys 2018 的最佳长论文(Best Long Paper Award)由Criteo 人工智能实验室的Stephen Bonner和 Flavian Vasile获得。他们提出因果嵌入的推荐方法。


最佳长论文Causal Embeddings for Recommendation

论文链接:

http://www.zhuanzhi.ai/paper/f85a6456c9f08b6584ca62c05279147d


许多当前的应用程序使用推荐来影响自然用户行为,例如增加销售数量或在网站上花费的时间。这导致最终推荐目标与经典设置之间的差距,其中推荐候选者的效果是通过其与过去用户行为的一致性来评估完成的,通过预测用户项矩阵中的缺失条目或最可能的下一事件来实现。为了弥补这一差距,我们优化了推荐策略,以便针对有机用户行为增加预期结果。我们证明这相当于学习在完全随机推荐策略下预测推荐结果。为此,我们提出了一种新的领域自适应算法,该算法从包含偏向推荐策略结果的记录数据中学习,并根据随机暴露预测推荐结果。我们将提出的方法与最先进的因子分解方法以及新的因果推荐方法进行比较,实验结果显示出显著的效果提升。


Many current applications use recommendations in order to modify the natural user behavior, such as to increase the number of sales or the time spent on a website. This results in a gap between the final recommendation objective and the classical setup where recommendation candidates are evaluated by their coherence with past user behavior, by predicting either the missing entries in the user-item matrix, or the most likely next event. To bridge this gap, we optimize a recommendation policy for the task of increasing the desired outcome versus the organic user behavior. We show this is equivalent to learning to predict recommendation outcomes under a fully random recommendation policy. To this end, we propose a new domain adaptation algorithm that learns from logged data containing outcomes from a biased recommendation policy and predicts recommendation outcomes according to random exposure. We compare our method against state-of-the-art factorization methods, in addition to new approaches of causal recommendation and show significant improvements.

此外,另一篇论文获得Best Long Paper Runner-up

Generation Meets Recommendation: Proposing Novel Items for Groups of Users
by T.V. Vo, H. Soh (Best Long Paper Runner-up)

论文地址

https://doi.org/10.1145/3240323.3240357



最佳短论文

RecSys 2018 的最佳短论文(Best Short Paper Award)由德国Duisburg-Essen大学的Benedikt Loepp等人获得。他们提出研究用户推荐评估中的物品消费的影响。


最佳短论文Impact of Item Consumption on Assessment of Recommendations in User Studies


论文链接:

https://doi.org/10.1145/3240323.3240375


在推荐系统的用户研究中,参与者通常不能消费推荐的物品。他们被要求通过问卷调查评估推荐质量和与用户体验相关的其他特性。然而,如果没有听过推荐的歌曲或看过建议的电影,这可能是一个容易出错的任务,可能会限制这些研究中获得的结果的有效性。在本文中,我们研究了实际消费推荐物品的效果。我们提出两个在不同领域进行的用户研究,在某些情况下表明,建议和问卷调查结果的评估存在差异。显然,在不允许用户使用物品的情况下,并不总是能够充分测量用户体验。另一方面,根据领域和提供的信息,参与者有时似乎相当接近推荐的实际价值。


In user studies of recommender systems, participants typically cannot consume the recommended items. Still, they are asked to assess recommendation quality and other aspects related to user experience by means of questionnaires. Without having listened to recommended songs or watched suggested movies, however, this might be an error-prone task, possibly limiting validity of results obtained in these studies. In this paper, we investigate the effect of actually consuming the recommended items. We present two user studies conducted in different domains showing that in some cases, differences in the assessment of recommendations and in questionnaire results occur. Apparently, it is not always possible to adequately measure user experience without allowing users to consume items. On the other hand, depending on domain and provided information, participants sometimes seem to approximate the actual value of recommendations reasonably well.


此外,另一篇论文获得Best Short Paper Runner-up

HOP-rec: High-order Proximity for Implicit Recommendation
by J.-H. Yang, C.-M. Chen, C.-J. Wang, M.-F. Tsai (Best Short Paper Runner-up)


论文地址

https://dl.acm.org/citation.cfm?doid=3240323.3240381



-END-

专 · 知


人工智能领域26个主题知识资料全集获取与加入专知人工智能服务群: 欢迎微信扫一扫加入专知人工智能知识星球群,获取专业知识教程视频资料和与专家交流咨询!


请PC登录www.zhuanzhi.ai或者点击阅读原文,注册登录专知,获取更多AI知识资料!


请加专知小助手微信(扫一扫如下二维码添加),加入专知主题群(请备注主题类型:AI、NLP、CV、 KG等)交流~

 AI 项目技术 & 商务合作:bd@zhuanzhi.ai, 或扫描上面二维码联系!

请关注专知公众号,获取人工智能的专业知识!

点击“阅读原文”,使用专知

展开全文
Top
微信扫码咨询专知VIP会员