Maximizing Sequence-Submodular Functions and Its Application to Online Advertising

• 成果类型: Article
• 署名作者: Alaei, Saeed; Makhdoumi, Ali; Malekian, Azarakhsh
• 署名单位: Alphabet Inc.; Google Incorporated; Duke University; University of Toronto
• 刊物名称: MANAGEMENT SCIENCE
• ISSN/ISSBN: 0025-1909
• DOI: 10.1287/mnsc.2020.3820
• 发表日期: 2021
• 页码: 6030-6054
• 关键词: submodular function maximization; sequence submodularity; applications to online advertisement; online ad allocation; query rewriting

摘要

Motivated by applications in online advertising, we consider a class of maximization problems where the objective is a function of the sequence of actions and the running duration of each action. For these problems, we introduce the concepts of sequence-submodularity and sequence-monotonicity, which extend the notions of submodularity and monotonicity from functions defined over sets to functions defined over sequences. We establish that if the objective function is sequence-submodular and sequence-nondecreasing, then there exists a greedy algorithm that achieves 1 - 1/e of the optimal solution. We apply our algorithm and analysis to two applications in online advertising: online ad allocation and query rewriting. We first show that both problems can be formulated as maximizing nondecreasing sequence-submodular functions. We then apply our framework to these two problems, leading to simple greedy approaches with guaranteed performances. In particular, for the online ad allocation problem, the performance of our algorithm is 1 - 1/e, which matches the best known existing performance, and for the query rewriting problem, the performance of our algorithm is 1 - 1/e1-1/e, which improves on the best known existing performance in the literature.