Skip to main content

Management Research Seminar Series with Dr. Xiao Liu

Friday, April 22, 2022, 3:00 pm – 4:00 pm

Management Research Seminar Series, conducted by the Department of Economics and Business, aims to bring accomplished researchers in the Management field to share their current research projects to facilitate an academic discussion, enhance knowledge, and discover potential connections. The series is designed for the academic audience, i.e., the Faculty and MA, PhD students; however, anyone interested in the series is welcome to attend. 

On April 22, Dr. Xiao Liu from New York University Stern School of Business will join the Management Research Seminar Series. The event will be on Zoom. For details and the zoom link, please email or

Title: Dynamic Coupon Targeting Using Batch Deep Reinforcement Learning: An Application to Livestream Shopping

Abstract: We present an empirical framework to create dynamic coupon targeting strategies for high-dimensional and high-frequency settings and test its performance using a large-scale field experiment. The framework captures consumers’ intertemporal tradeoffs associated with dynamic pricing. It does not rely on functional form assumptions about consumers’ decision-making processes. The model is estimated using batch deep reinforcement learning (BDRL), which relies on Q-learning, a model-free solution that can mitigate model bias. It leverages deep neural networks to represent the high-dimensional state space and alleviate the curse of dimensionality. It allows policy learning and policy evaluation to operate in the batch mode to safeguard against monetary losses. The empirical application is in a multi-billion-dollar livestream shopping context where hosts showcase products and interact with consumers in real-time. Our BDRL solution is twice as effective as static targeting policies, and 20% more effective than the model-based solution in increasing the platform’s revenue. Our recommended targeting strategy involves cross-sectional and intertemporal price discrimination and their interaction. Cross-sectionally, we recommend giving lower discounts when consumers visit more attractive hosts. Intertemporally, we recommend price skimming. Combining the two, we recommend increasing the coupon discount level at a faster rate for low spenders than for high spenders.