{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/online-influence-maximization-in-non","title":"Online Influence Maximization in Non-Stationary Social Networks","arxiv_id":"1604.07638","date":"2016-04-26","proceeding":null,"authors":["Yixin Bao","Xiaoke Wang","Zhi Wang","Chuan Wu","Francis C. M. Lau"],"abstract":"Social networks have been popular platforms for information propagation. An\nimportant use case is viral marketing: given a promotion budget, an advertiser\ncan choose some influential users as the seed set and provide them free or\ndiscounted sample products; in this way, the advertiser hopes to increase the\npopularity of the product in the users' friend circles by the world-of-mouth\neffect, and thus maximizes the number of users that information of the\nproduction can reach. There has been a body of literature studying the\ninfluence maximization problem. Nevertheless, the existing studies mostly\ninvestigate the problem on a one-off basis, assuming fixed known influence\nprobabilities among users, or the knowledge of the exact social network\ntopology. In practice, the social network topology and the influence\nprobabilities are typically unknown to the advertiser, which can be varying\nover time, i.e., in cases of newly established, strengthened or weakened social\nties. In this paper, we focus on a dynamic non-stationary social network and\ndesign a randomized algorithm, RSB, based on multi-armed bandit optimization,\nto maximize influence propagation over time. The algorithm produces a sequence\nof online decisions and calibrates its explore-exploit strategy utilizing\noutcomes of previous decisions. It is rigorously proven to achieve an\nupper-bounded regret in reward and applicable to large-scale social networks.\nPractical effectiveness of the algorithm is evaluated using both synthetic and\nreal-world datasets, which demonstrates that our algorithm outperforms previous\nstationary methods under non-stationary conditions.","url_abs":"http://arxiv.org/abs/1604.07638v1","url_pdf":"http://arxiv.org/pdf/1604.07638v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"online-influence-maximization-in-non","repo_url":"https://github.com/olety/TIMLinUCB","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"marketing","task_name":"Marketing"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}