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 2411.01410cs.LGcs.AIcs.SI
收藏

面向链接预测的PageRank Bandits

PageRank Bandits for Link Prediction

University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校);Meta AI;University of Rochester(罗切斯特大学)

Yikun Ban, Jiaru Zou, Zihao Li, Yunzhe Qi, Dongqi Fu, Jian Kang, Hanghang Tong, Jingrui He

AI总结:

针对传统监督式链接预测难以适配动态兴趣、难平衡利用与探索的问题,提出融合上下文老虎机与PageRank的PRB算法,设计新奖励并提供理论保证,在线离线实验均验证了其有效性。

摘要:

链接预测是图学习中的关键问题,在推荐系统、知识图谱补全等领域有广泛应用。已有大量研究致力于解决该问题,包括基于相似度指标的方法和图神经网络(Graph Neural Networks, GNN)方法。但现有多数方案仍以传统监督学习为基础,难以随时间推移适配不断变化的用户兴趣,也无法解决链接预测中固有的利用与探索两难困境。针对这些挑战,本文将链接预测重新建模为序列决策过程,每次链接预测交互按顺序发生。我们提出了一种新型融合算法PRB(PageRank Bandits,即PageRank老虎机),这是首个将上下文老虎机与PageRank结合以实现协同利用与探索的算法。我们还提出了一种新的奖励构造方式,并为PRB提供了理论性能保证。最后,我们在在线和离线两种设置下对PRB进行了广泛评估,将其与基于老虎机和基于图的方法进行对比。PRB的实证成功证明了所提出的融合方法的价值。我们的代码已发布在https://github.com/jiaruzouu/PRB。

英文摘要:

Link prediction is a critical problem in graph learning with broad applications such as recommender systems and knowledge graph completion. Numerous research efforts have been directed at solving this problem, including approaches based on similarity metrics and Graph Neural Networks (GNN). However, most existing solutions are still rooted in conventional supervised learning, which makes it challenging to adapt over time to changing customer interests and to address the inherent dilemma of exploitation versus exploration in link prediction. To tackle these challenges, this paper reformulates link prediction as a sequential decision-making process, where each link prediction interaction occurs sequentially. We propose a novel fusion algorithm, PRB (PageRank Bandits), which is the first to combine contextual bandits with PageRank for collaborative exploitation and exploration. We also introduce a new reward formulation and provide a theoretical performance guarantee for PRB. Finally, we extensively evaluate PRB in both online and offline settings, comparing it with bandit-based and graph-based methods. The empirical success of PRB demonstrates the value of the proposed fusion approach. Our code is released at https://github.com/jiaruzouu/PRB.

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。

↑