Yi-Chun Akchen (UCL)

I am an Assistant Professor at the UCL School of Management, University College London, since August 2023. My research aims to understand customer choice from a data-driven perspective and provides operational insights using machine learning and optimization techniques.

I received my PhD from the UCLA Anderson School of Management, where I was fortunate to be advised by Professor Velibor Mišić. I received my master’s degree from Stanford University and bachelor’s degree from National Taiwan University.

Working Papers

Consider or Choose? The Role and Power of Consideration Sets
with Dmitry Mitrofanov
Major revision in Management Science

On Size Substitution and Its Role in Assortment and Inventory Planning
with Felipe Caro
Major revision in Manufacturing & Service Operations Management
Assortment Optimization under the Decision Forest Model
with Velibor Mišić
Major revision in Manufacturing & Service Operations Management

Refereed Journal Papers

Column-Randomized Linear Programs: Performance Guarantees and Applications
with Velibor Mišić
Forthcoming in Operations Research (2024+) [DOI]
Decision Forest: A Nonparametric Approach to Modeling Irrational Choice
with Velibor Mišić
Management Science (2022) [MS Blog] [DOI] [Code]
- Honorable Mention, George Nicholson Student Paper Prize, INFORMS, 2020.
- 1st Place, Decision Analysis Student Paper Award, INFORMS, 2019.
- 2nd Place, Jeff McGill Revenue Management and Pricing Student Paper Award, INFORMS, 2019.
- Finalist, Service Science Best Paper Award, INFORMS, 2019.
- Spotlight Presentation, 2019 INFORMS Revenue Management and Pricing Conference.

Deep Reinforcement Learning for Event-Driven Multi-Agent Decision Processes
with Kunal Menda, Justin Grana, James Bono, Brendan Tracey, Mykel Kochenderfer, David Wolpert
IEEE Transactions on Intelligent Transportation Systems (2019)
Learning Discrete Bayesian Networks from Continuous Data
with Tim Wheeler and Mykel Kochenderfer
Journal of Artificial Intelligence Research (2017)

Refereed Conference Papers

Improving Offline Value-Function Approximations for POMDPs by Reducing Discount Factors
with Mykel Kochenderfer and Matthijs Spaan
IEEE/RSJ International Conference on Intelligent Robots and Systems (2018)

Teaching

Statistics for Business Research
MSc Management program, UCL School of Management (Fall, 2023 - 2024).