Evaluating Deep Vs. Wide & Deep Learners As Contextual Bandits For Personalized Email Promo Recommendations
This work addresses personalized email promotion recommendations for businesses, but it is incremental as it compares existing methods on synthetic data.
The paper tackled the problem of predicting optimal promotional offers for customers using contextual bandits, finding that wide and deep networks and deep-only networks with embeddings achieve similar prediction accuracies on a synthetic email promo dataset, with slight performance improvements from using upper confidence bound or Thompson sampling via Monte Carlo dropout.
Personalization enables businesses to learn customer preferences from past interactions and thus to target individual customers with more relevant content. We consider the problem of predicting the optimal promotional offer for a given customer out of several options as a contextual bandit problem. Identifying information for the customer and/or the campaign can be used to deduce unknown customer/campaign features that improve optimal offer prediction. Using a generated synthetic email promo dataset, we demonstrate similar prediction accuracies for (a) a wide and deep network that takes identifying information (or other categorical features) as input to the wide part and (b) a deep-only neural network that includes embeddings of categorical features in the input. Improvements in accuracy from including categorical features depends on the variability of the unknown numerical features for each category. We also show that selecting options using upper confidence bound or Thompson sampling, approximated via Monte Carlo dropout layers in the wide and deep models, slightly improves model performance.