We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.IR

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Information Retrieval

Title: The Use of Bandit Algorithms in Intelligent Interactive Recommender Systems

Authors: Qing Wang
Abstract: In today's business marketplace, many high-tech Internet enterprises constantly explore innovative ways to provide optimal online user experiences for gaining competitive advantages. The great needs of developing intelligent interactive recommendation systems are indicated, which could sequentially suggest users the most proper items by accurately predicting their preferences, while receiving the up-to-date feedback to refine the recommendation results, continuously. Multi-armed bandit algorithms, which have been widely applied into various online systems, are quite capable of delivering such efficient recommendation services. However, few existing bandit models are able to adapt to new changes introduced by the modern recommender systems.
Comments: 10 pages
Subjects: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI)
Cite as: arXiv:2107.00161 [cs.IR]
  (or arXiv:2107.00161v1 [cs.IR] for this version)

Submission history

From: Qing Wang [view email]
[v1] Thu, 1 Jul 2021 00:43:05 GMT (446kb,D)

Link back to: arXiv, form interface, contact.