arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2302.03431cs.IR
收藏

推荐中潜在行为空间的挖掘与规则化

Exploration and Regularization of the Latent Action Space in Recommendation

Shuchang Liu, Qingpeng Cai, Bowen Sun, Yuhao Wang, Ji Jiang, Dong Zheng, Kun Gai, Peng Jiang, Xiangyu Zhao, Yongfeng Zhang

摘要:

在推荐系统中,强化学习解决方案有效地提高了推荐性能,因为它们能够捕获长期的用户-系统交互。然而,推荐策略的动作空间是一个项目列表,对于动态候选项目池,该列表可能非常大。为了克服这一挑战,我们提出了一个超级演员和评论家的学习框架,该政策分解的项目列表生成过程中的超动作推理步骤和效果动作选择步骤。第一步将给定的状态空间映射到向量化的超动作空间,第二步基于超动作选择项目列表。为了调节两个动作空间之间的差异,我们设计了一个对齐模块,以及一个核映射函数的项目,以确保推理的准确性,并包括一个监督模块,以稳定的学习过程。我们在公共数据集上构建模拟环境,并根据经验表明,与标准RL基线相比,我们的框架在推荐方面更优越。

英文摘要:

In recommender systems, reinforcement learning solutions have effectively boosted recommendation performance because of their ability to capture long-term user-system interaction. However, the action space of the recommendation policy is a list of items, which could be extremely large with a dynamic candidate item pool. To overcome this challenge, we propose a hyper-actor and critic learning framework where the policy decomposes the item list generation process into a hyper-action inference step and an effect-action selection step. The first step maps the given state space into a vectorized hyper-action space, and the second step selects the item list based on the hyper-action. In order to regulate the discrepancy between the two action spaces, we design an alignment module along with a kernel mapping function for items to ensure inference accuracy and include a supervision module to stabilize the learning process. We build simulated environments on public datasets and empirically show that our framework is superior in recommendation compared to standard RL baselines.

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。

↑