My research interests are in the interface of algorithm design under uncertainty and the operations of digital platforms. I try to better understand how to allocate resources and personalize assortment offerings under various forms of uncertainty, e.g., stochastic choice preferences and dynamic networks. Lately, I am passionate about deploying similar data-driven approaches in cultural institutions to help reach broader audiences and experiment with new curation methods.
Before joining LBS, I was a data scientist in the Marketplace Optimization group at Uber Technologies in 2017-2018. I worked and consulted for the Matching Science team to design and experiment with new matching and pricing algorithms. I received a PhD in Operations Research from Massachusetts Institute of Technology (MIT) in 2017, where I was advised by and collaborated with Profs. Vivek Farias, Retsef Levi, and Danny Segev. Before MIT, I earned an MS in Applied Mathematics from the Ecole Polytechnique (Paris) in 2013. I was born in Meknes, Morocco.
keywords: dynamic matching, choice modeling, assortment and inventory optimization, digital platforms, cultural institutions, approximation algorithms.
A Nonparametric Framework for Online Stochastic Matching with Correlated Arrivals, A. and Ma, Working paper (2022)
- Extended abstract accepted in The 21st ACM Conference on Economics and Computation (EC), 2020
choice modelling via machine learning
Representing Random Utility Choice Models with Neural Networks, A. and Desir, Under revision (2022) [code]
assortment & platform design, applications
Designing Layouts for Sequential Experiences: Application to Cultural Institutions, A., Deshmane, and Martínez-de-Albéniz, Working paper (2022)
Algorithmic Collusion in Assortment Games, A. and den Boer, Under revision (2021)
- Accepted in the EC 2021 Workshop on the Design of Online Platforms: Frontiers and Challenges
Online Assortment Optimization for Two-sided Matching Platforms, A. and Saban, Forthcoming in Management Science (2022)
- Extended abstract accepted in The 22nd ACM Conference on Economics and Computation (EC), 2021
- Spotlight track, Revenue Management and Pricing Conference, 2021
The Click-Based MNL Model: A Novel Framework for Modeling Click Data in Assortment Optimization, A., Feldman and Segev, Under review (2022)
- Spotlight track, Revenue Management and Pricing Conference, 2019
The Exponomial Choice Model for Assortment Optimization: An Alternative to the MNL Model?, A., Feldman and Segev, Forthcoming in Management Science (2022)
Display Optimization for Vertically Differentiated Locations Under Multinomial Logit Preferences, A. and Segev, Management Science (2020)
Assortment Optimization Under Consider-then-Choose Choice Models, A., Farias and Levi, Management Science (2020)
inventory management with substitution
Greedy-Like Algorithms for Dynamic Assortment Planning Under Multinomial Logit Preferences, A., Levi and Segev, Operations Research (2018)[code]
- Finalist in the 2021 MS&OM Best OM Paper Published in Operations Research
- Finalist in the 2016 INFORMS Student Paper Nicholson Prize
Approximation Algorithms for Dynamic Assortment Optimization Models, A., Levi and Segev, Mathematics of Operations Research (2018) [code]
The Stability of MNL-Based Demand under Dynamic Customer Substitution and its Algorithmic Implications, A. and Segev, Under review (2020) [code]
approximation algorithms misc.
The Ordered k-Median Problem: Surrogate Models and Approximation Algorithms, A. and Segev, Mathematical Programming (2018)
Technical Note -- An Approximate Dynamic Programming Approach for the Incremental Knapsack Problem, A. and Segev, Forthcoming in Operations Research (2022)
Technical Note -- The Approximability of Assortment Optimization Under Ranking Preferences, A., Farias, Levi and Segev, Operations Research (2018)