In many digital contexts such as online news and e-tailing with many new users and items, recommendation systems face several challenges: i) how to make initial recommendations to users with little or no response history (i.e., cold-start problem), ii) how to learn user preferences on items (test and learn), and iii) how to scale across many users and items with myriad demographics and attributes. While many recommendation systems accommodate aspects of these challenges, few if any address all. This paper introduces a Collaborative Filtering (CF) Multi-armed Bandit (B) with Attributes (A) recommendation system (CFB-A) to jointly accommodate all of these considerations. Empirical applications including an offline test on MovieLens data, synthetic data simulations, and an online grocery experiment indicate the CFB-A leads to substantial improvement on cumulative average rewards (e.g., total money or time spent, clicks, purchased quantities, average ratings, etc.) relative to the most powerful extant baseline methods.

翻译：在许多数字场景中，例如在线新闻和电子商务中，由于存在大量新用户和新物品，推荐系统面临若干挑战：i）如何向几乎没有或完全没有响应历史记录的用户进行初始推荐（即冷启动问题），ii）如何学习用户对物品的偏好（测试与学习），以及iii）如何在具有众多人口统计特征和属性的海量用户与物品之间实现扩展。尽管许多推荐系统能应对这些挑战的某些方面，但几乎没有一个系统能同时解决所有问题。本文提出一种基于协同过滤（CF）的多臂赌博机（B）并融合属性（A）的推荐系统（CFB-A），以联合处理以上所有考量因素。实证应用包括基于MovieLens数据的离线测试、合成数据模拟以及一项在线杂货实验，结果表明，与现有最强大的基线方法相比，CFB-A在累积平均奖励（例如总消费金额或时间、点击量、购买数量、平均评分等）上实现了显著提升。