NYC Gazette

Researchers at the Katz School's Graduate Department of Computer Science and Engineering are utilizing advanced artificial intelligence tools to enhance customer targeting strategies. In a study titled "Optimizing Customer Targeting Using Reinforcement Learning and Neural Networks for Adaptive Marketing Strategies," set to be introduced at the 2025 IEEE International Conference on Consumer Electronics, the team presents an innovative framework employing reinforcement learning (RL).

Dr. David Li, senior author of the study and program director of the M.S. in Data Analytics and Visualization, explained that the framework centers around two key components: "the mean-stat strategy and a neural network-based approach using the cross-entropy method." These elements enable businesses to refine their marketing efforts more effectively.

The RL model simulates interactions between businesses and customers, aiming to identify high-value customers by modeling their spending potential as a normal distribution. Dr. Li noted that this approach allows for balancing exploration of new strategies with exploitation of identified valuable customers through statistical confidence intervals rather than fixed probabilities.

Lead author Zubair Khan, a student in the M.S. in Artificial Intelligence program, emphasized that this "data-driven approach ensures better targeting while avoiding wasted effort on low-value customers." The neural networks within this framework help uncover complex patterns in customer behavior by analyzing extensive datasets.

In practical terms, if a business observes positive responses from high-value customers to specific promotions, the neural network integrates these findings into its strategy. This method is applicable across various industries such as e-commerce and hospitality, where targeted promotions can lead to increased returns.

Dr. Li reported that simulations conducted with this framework demonstrated superior performance over traditional methods like ϵ-greedy in identifying high-value customers. The strategies achieved higher cumulative rewards and adapted well to changing customer behaviors.

"Our study sets a strong foundation for the future of dynamic marketing," said Khan, highlighting how advanced statistical techniques combined with machine learning offer businesses enhanced capabilities in understanding customer behaviors.