Jinqiu Deng
Director of Algorithms, JD
He has many years of experience in Silicon Valley and Wall Street, and has worked as an algorithm expert for Uber and Bloomberg, focusing on the research and application of supply chain management, pricing strategy optimization, dynamic pricing, causal inference, and large language models. He is now the Director of Retail Algorithms at Jingdong, responsible for supply chain management strategy optimization, pricing algorithm optimization, price strategy development, and price ecosystem governance. His team won the 2024 INFORMS Prize, the first Asian team to win the prize in the 34 years of its existence.
Topic
Innovative Inventory Liquidation Practices Based on Causal Grand Modeling
Introduction: In recent years, the combination of causal inference and grand modeling has become a cutting-edge technique to drive supply chain management innovation, and has shown great potential especially in inventory liquidation. This approach enables inventory liquidation to be more accurate and efficient by accurately analyzing causal relationships. In this talk, we will discuss how to combine causal inference with big model theory to solve the inventory clearance problem, including accurate identification of target products, reasonable setting of clearance cycle, and optimization of inventory management strategy. At the same time, we will share the data processing challenges faced in practical applications, the technical difficulties in model landing, and the specific application effects. Outline: 1、Business Background: Challenges and Opportunities of Inventory Clearance 2、program selection: causal inference and the advantages of the combination of large models 3、landing challenges: data quality, model training and real-time optimization issues 4、Solution Ideas: Algorithm Optimization Strategies and Technical Implementation Paths 5、Industry outlook: further application of big models in supply chain management
