Research

Research papers

[1] Heterogeneous Values of One Data Point: Active Label Acquisition in Assortment Optimization
Mo Liu, Junyu Cao, Zuo-jun Max Shen. To be resubmitted to Management Science. [link] [Slides]

• “How much is the marginal contribution to the revenue increase when adding one new customer with some specific feature into the training set?” (Job market project)

• By evaluating this marginal contribution of each customer, we are able to identify informative customers and reduce the size of the training set by about 80% percent while maintaining the same level of revenue.

[2] Active Learning in the Predict-then-Optimize Framework: A Margin-Based Approach
Mo Liu, Paul Grigas, Heyuan Liu, Zuo-jun Max Shen. To be resubmitted to Management Science. [link] [Slides]

• “How to identify informative samples for the decision-making?” (Job market project)
• Best Student Paper Nominee at INFORMS Workshop on Data Science 2023
• Second Place Poster Prize at YinzOR 2023

[3] Learning from Click Transition Data: The Effectiveness of Greedy Pricing Policy under Dynamic Product Availability
Mo Liu, Junyu Cao, Zuo-jun Max Shen. To be submitted to Management Science. [link] , [5-minute Talk] (Previous version) [Slides] (Previous version)

• Finalist at 2023 INFORMS Service Science Student Competition
• Fan Favoriate Flash Talk at YinzOR 2023

[4] Active Learning by Prediction Uncertainty: Applications in Contextual Stochastic Linear Optimization
Mo Liu, Paul Grigas, Zuo-jun Max Shen. Woking paper. [link]

[5] End-to-End Deep Learning for Automatic Inventory Management with Fixed Ordering Cost
Mo Liu, Meng Qi, Zuo-jun Max Shen. Working paper. [link]

Patent in application

Machine Learning And Optimization With Partially Observable Time Series Data
Zachary Xue, Mo Liu, Markus Ettl, Shivaram Subramanian

Selected Talks

• INFORMS Workshop on Data Science 2023, Active Learning in the Predict-then-Optimize Framework: A Margin-Based Approach, Oct 14, 2023
• INFORMS Service Science Student Competition 2023, Pricing under the Generalized Markov Chain Choice Model: Learning through Large-scale Click Behaviors, October 16
• INFORMS Annual Meeting Talk, Active Label Acquisition with Personalized Incentives in Assortment Optimization, Oct 17
• Purdue Operations Conference, Active Label Acquisition with Personalized Incentives in the Assortment Optimization, Sept. 9, 2023
• CMU YinzOR Student Competition, Pricing under the Generalized Markov Chain Choice Model: Learning through Large-scale Click Behaviors, August, 2023
• International Conference Stochastic Programming, Active Learning in the Predict-then-Optimize Framework: A Margin-Based Approach, July, 2023
• MSOM Conference, Personalized Incentive for Active Label Acquisition in the Assortment Optimization, June, 2023
• IBM Research Intern Talk, Active Learning in the Predict-then-Optimize Framework: A Margin-Based Approach, June, 2022
• INFORMS Annual Meeting, Pricing under the Generalized Markov Chain Choice Model: Learning through Large-scale Click Behavior, Oct, 2022
• INFORMS Annual Meeting, End-to-End Deep Learning for the Inventory Management with Fixed Ordering Cost, Oct, 2020