The MNL-Bandit Problem

Shipra Agrawal*, Vashist Avadhanula, Vineet Goyal, Assaf Zeevi

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review


In this chapter, we consider a dynamic assortment optimization problem where customers choose according to an unknown MNL choice model. In each period, we offer a subset (assortment) of products and observe the customer’s choice to learn the model parameters. We present two dynamic assortment selection approaches that optimally balances the exploration–exploitation trade-off and does not require any prior knowledge about the model parameters.

Original languageEnglish
Title of host publicationSpringer Series in Supply Chain Management
PublisherSpringer Nature
Number of pages30
StatePublished - 2022
Externally publishedYes

Publication series

NameSpringer Series in Supply Chain Management
ISSN (Print)2365-6395
ISSN (Electronic)2365-6409


FundersFunder number
Columbia University


    Dive into the research topics of 'The MNL-Bandit Problem'. Together they form a unique fingerprint.

    Cite this