MNL-bandit: a dynamic learning approach to assortment selection
From MaRDI portal
Publication:5129205
DOI10.1287/OPRE.2018.1832zbMATH Open1444.90021arXiv1706.03880OpenAlexW2972989214WikidataQ127280865 ScholiaQ127280865MaRDI QIDQ5129205FDOQ5129205
Authors: Shipra Agrawal, Vashist Avadhanula, Vineet Goyal, Assaf Zeevi 
Publication date: 26 October 2020
Published in: Operations Research (Search for Journal in Brave)
Abstract: We consider a dynamic assortment selection problem, where in every round the retailer offers a subset (assortment) of substitutable products to a consumer, who selects one of these products according to a multinomial logit (MNL) choice model. The retailer observes this choice and the objective is to dynamically learn the model parameters, while optimizing cumulative revenues over a selling horizon of length . We refer to this exploration-exploitation formulation as the MNL-Bandit problem. Existing methods for this problem follow an "explore-then-exploit" approach, which estimate parameters to a desired accuracy and then, treating these estimates as if they are the correct parameter values, offers the optimal assortment based on these estimates. These approaches require certain a priori knowledge of "separability", determined by the true parameters of the underlying MNL model, and this in turn is critical in determining the length of the exploration period. (Separability refers to the distinguishability of the true optimal assortment from the other sub-optimal alternatives.) In this paper, we give an efficient algorithm that simultaneously explores and exploits, achieving performance independent of the underlying parameters. The algorithm can be implemented in a fully online manner, without knowledge of the horizon length . Furthermore, the algorithm is adaptive in the sense that its performance is near-optimal in both the "well separated" case, as well as the general parameter setting where this separation need not hold.
Full work available at URL: https://arxiv.org/abs/1706.03880
Recommendations
- Optimal policy for dynamic assortment planning under multinomial logit models
- Dynamic assortment optimization with a multinomial logit choice model and capacity constraint
- Greedy-like algorithms for dynamic assortment planning under multinomial logit preferences
- Dynamic assortment personalization in high dimensions
- Learning consumer tastes through dynamic assortments
Consumer behavior, demand theory (91B42) Transportation, logistics and supply chain management (90B06)
Cites Work
- Title not available (Why is that?)
- Asymptotically efficient adaptive allocation rules
- Discrete Choice Methods with Simulation
- Some aspects of the sequential design of experiments
- Finite-time analysis of the multiarmed bandit problem
- Assortment optimization under variants of the nested logit model
- Dynamic assortment optimization with a multinomial logit choice model and capacity constraint
- Revenue Management Under a General Discrete Choice Model of Consumer Behavior
- Title not available (Why is that?)
- 10.1162/153244303321897663
- Title not available (Why is that?)
- Probability and Computing
- Regret analysis of stochastic and nonstochastic multi-armed bandit problems
- A General Attraction Model and Sales-Based Linear Program for Network Revenue Management Under Customer Choice
- Demand Estimation and Assortment Optimization Under Substitution: Methodology and Application
- Dynamic assortment with demand learning for seasonal consumer goods
- The d-Level Nested Logit Model: Assortment and Price Optimization Problems
- A Markov chain approximation to choice modeling
- Title not available (Why is that?)
- Linearly parameterized bandits
- On the tightness of an LP relaxation for rational optimization and its applications
- Near-Optimal Regret Bounds for Thompson Sampling
- A note on a tight lower bound for capacitated MNL-bandit assortment selection models
Cited In (18)
- Game of Thrones: Fully Distributed Learning for Multiplayer Bandits
- Optimal Policy for Dynamic Assortment Planning Under Multinomial Logit Models
- A tractable online learning algorithm for the multinomial logit contextual bandit
- Stochastic approximation for uncapacitated assortment optimization under the multinomial logit model
- Title not available (Why is that?)
- Transfer learning for contextual multi-armed bandits
- Learning consumer tastes through dynamic assortments
- Title not available (Why is that?)
- Optimal pricing of online products based on customer anchoring‐adjustment psychology
- Dynamic Assortment Personalization in High Dimensions
- Continuous Assortment Optimization with Logit Choice Probabilities and Incomplete Information
- A regret lower bound for assortment optimization under the capacitated MNL model with arbitrary revenue parameters
- Smoothness-Adaptive Contextual Bandits
- Dynamic assortment with demand learning for seasonal consumer goods
- Assortment optimization: a systematic literature review
- A Learning Approach for Interactive Marketing to a Customer Segment
- Robust Learning of Consumer Preferences
- Stochastic continuum-armed bandits with additive models: minimax regrets and adaptive algorithm
Uses Software
This page was built for publication: MNL-bandit: a dynamic learning approach to assortment selection
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5129205)
