MultiSlot ReRanker：推荐系统中一种通用的基于模型的重排序框架
MultiSlot ReRanker: A Generic Model-based Re-Ranking Framework in Recommendation Systems
AI总结：
本文提出通用多目标重排序框架 MultiSlot ReRanker，通过顺序贪心算法和帕累托最优实现相关性、多样性与新鲜度的联合优化，并构建可配置模拟器以高效评估不同学习算法。
摘要：
在本文中，我们提出了一个通用的基于模型的重排序框架 MultiSlot ReRanker，它同时优化相关性、多样性和新鲜度。具体来说，我们的顺序贪心算法（SGA）具有足够高的效率（线性时间复杂度），适用于大规模生产级推荐引擎。它实现了离线接收者操作特征曲线下面积（AUC）$+6\%$ 到 $+10\%$ 的提升，这主要得益于显式建模列表中项目之间的相互影响，并利用多个目标的第二轮排序分数。此外，我们将离线回放理论推广到多槽位重排序场景，并在多个目标之间进行权衡。离线回放结果可以通过帕累托最优性进一步改进。而且，我们基于 OpenAI Gym 构建了一个多槽位重排序模拟器，并与 Ray 框架集成。它可以方便地针对不同假设进行配置，以快速对强化学习和监督学习算法进行基准测试。
英文摘要：
In this paper, we propose a generic model-based re-ranking framework, MultiSlot ReRanker, which simultaneously optimizes relevance, diversity, and freshness. Specifically, our Sequential Greedy Algorithm (SGA) is efficient enough (linear time complexity) for large-scale production recommendation engines. It achieved a lift of $+6\%$ to $ +10\%$ offline Area Under the receiver operating characteristic Curve (AUC) which is mainly due to explicitly modeling mutual influences among items of a list, and leveraging the second pass ranking scores of multiple objectives. In addition, we have generalized the offline replay theory to multi-slot re-ranking scenarios, with trade-offs among multiple objectives. The offline replay results can be further improved by Pareto Optimality. Moreover, we've built a multi-slot re-ranking simulator based on OpenAI Gym integrated with the Ray framework. It can be easily configured for different assumptions to quickly benchmark both reinforcement learning and supervised learning algorithms.
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。
