arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1609.01508cs.LG
收藏

第1001章潜伏混合的低阶盗贼

Low-rank Bandits with Latent Mixtures

Indian Institute of Science(印度科学理工学院);Inria Saclay – Île de France(法国国家信息与自动化研究所萨克雷-法兰西岛研究中心)

Aditya Gopalan, Odalric-Ambrym Maillard, Mohammadi Zaki

摘要:

我们研究的任务,最大限度地提高回报,从推荐项目(行动),用户顺序交互的推荐系统。用户被建模为C许多代表性的用户类的潜在混合物,其中每个类指定跨动作的平均奖励配置文件。用户特征(类上的混合分布)和项目特征(每个类的平均奖励向量)都是先验未知的。用户身份是学习者在交互时唯一可用的上下文信息。这在来自向用户b推荐项目a的预期回报ra,b的矩阵上引起低秩结构。当用户或物品侧特征完全已知时,问题简化为众所周知的线性强盗。在设置中,每个用户,其随机采样的口味配置文件,交互只有少量的会话,我们开发了一个强盗算法的双边不确定性。它结合了Anandkumar等人(2014 b)的鲁棒张量幂方法和Abbasi-Yadkori等人(2011)的OFUL线性强盗算法。我们提供了这个组合的第一个严格的遗憾分析,表明它的遗憾后,T用户交互是$\tilde O(C\sqrt{BT})$,与B的用户数量。对这一结果的一个成分是一个新的鲁棒性的OFUL,独立的利益。

英文摘要:

We study the task of maximizing rewards from recommending items (actions) to users sequentially interacting with a recommender system. Users are modeled as latent mixtures of C many representative user classes, where each class specifies a mean reward profile across actions. Both the user features (mixture distribution over classes) and the item features (mean reward vector per class) are unknown a priori. The user identity is the only contextual information available to the learner while interacting. This induces a low-rank structure on the matrix of expected rewards r a,b from recommending item a to user b. The problem reduces to the well-known linear bandit when either user or item-side features are perfectly known. In the setting where each user, with its stochastically sampled taste profile, interacts only for a small number of sessions, we develop a bandit algorithm for the two-sided uncertainty. It combines the Robust Tensor Power Method of Anandkumar et al. (2014b) with the OFUL linear bandit algorithm of Abbasi-Yadkori et al. (2011). We provide the first rigorous regret analysis of this combination, showing that its regret after T user interactions is $\tilde O(C\sqrt{BT})$, with B the number of users. An ingredient towards this result is a novel robustness property of OFUL, of independent interest.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑