题　目：Conditional Generative Learning for Joint Pricing and Inventory Control

嘉　宾：Xun Zhang, Assistant Professor, Southern University of Science and Technology

主持人：唐卓栋 助理教授 上海交通大学安泰经济与管理学院

时　间：2025年11月27日（周四）14:00-15:30

地　点：安泰楼A507室

内容简介:

This work considers data-driven joint pricing and inventory control with contextual information, in which the retailer makes pricing and inventory decisions based solely on historical data consisting of demand, price and covariates. Learning the optimal policy as a function of the covariates is challenging, as it requires knowledge of the conditional demand distribution, which is difficult to estimate and varies with the pricing decision. I propose a provably efficient deep conditional generative framework to learn the demand distribution conditional on the selling price and covariates from historical data. The conditional demand estimator is a deep generator driven by price, covariates and an easy-to-sample noise variable independent of both. This generator explicitly encodes the indirect effect of the pricing decision on demand, and it allows the data-driven control to be reformulated as a stochastic program and solved via stochastic gradient descent. I establish finite-sample error bounds for the excess cost or risk of the data-driven policy relative to the ground-truth optimal policy, as well as for the estimation error of the cost or risk estimator. The finite-sample bounds provide managers with insights into the reliability of the data-driven policy, and the optimal cost estimator gives a clear assessment of profitability when adopting the policy. Although the policy is learned from observational data, our sensitivity analysis suggests that confounding biases may have a limited impact on its performance. Comprehensive simulations show that the proposed approach is markedly faster than the state-of-the-art prescriptive benchmarks and achieves a significantly lower expected cost. A case study on real meal delivery data further confirms that our method is at least four times faster than the machine learning benchmarks, with at least a 10% reduction in out-of-sample costs. 

演讲人简介：

Xun Zhang is an Assistant Professor at the College of Business, Southern University of Science and Technology. He earned his B.S. in Statistics from Nanjing University in 2017 and his Ph.D. in Industrial Systems Engineering and Management from the National University of Singapore in 2021, advised by Dr. Zhisheng Ye and Dr. William B. Haskell. He then served as a Research Fellow at NUS (2021–2024) under the supervision of Dr. Zhisheng Ye and Prof. Teo Chung Piaw. He was awarded as the National-Level Excellent Young Scholar of China.

His research develops novel, data-driven methodologies to address core challenges at the intersection of operations management, machine learning, and statistics. In particular, his research program centers on three interconnected thrusts: supply chain and inventory management, online resource allocation, and large-scale optimization algorithms. For more information, please see his personal website: https://www.xunzhangweb.com/

欢迎广大师生参加！