Dynamic Personalization with Multiple Customer Signals: Multi-Response State Representation in Reinforcement Learning – MSI

Working Paper

Dynamic Personalization with Multiple Customer Signals: Multi-Response State Representation in Reinforcement Learning

Liangzong Ma, Ta-Wei Huang, Eva Ascarza and Ayelet Israeli

Mar 21, 2025

Download

Demonstrates a novel machine learning method that can be applied to “noisy” first-party data typically focused on short-term effects to optimize long-term outcomes such as customer lifetime value (CLV).

Topics

research recap

Beyond Next Best Actions: Optimizing Short-Term Interventions for Long-Term Value

Reinforcement machine learning can help companies understand their customers, but first-party data is typically noisy and focused on short-term effects that may not optimize long-term outcomes such as customer lifetime value (CLV). A novel Multi-Response State Representation (MRSR) Learning method can address these challenges to go beyond next best actions.

By using MSI.org you agree to our use of cookies as identifiers and for other features of the site as described in our Privacy Policy.