Learning Continuous Treatment Policy and Bipartite Embeddings for Matching with Heterogeneous Causal Effects

Causal inference methods are widely applied in the fields of medicine, policy, and economics. Central to these applications is the estimation of treatment effects to make decisions. Current methods make binary yes-or-no decisions based on the treatment effect of a single outcome dimension. These methods are unable to capture continuous space treatment policies with a measure of intensity. They also lack the capacity to consider the complexity of treatment such as matching candidate treatments with the subject. We propose to formulate the effectiveness of treatment as a parametrizable model, expanding to a multitude of treatment intensities and complexities through the continuous policy treatment function, and the likelihood of matching. Our proposal to decompose treatment effect functions into effectiveness factors presents a framework to model a rich space of actions using causal inference. We utilize deep learning to optimize the desired holistic metric space instead of predicting single-dimensional treatment counterfactual. This approach employs a population-wide effectiveness measure and significantly improves the overall effectiveness of the model. The performance of our algorithms is. demonstrated with experiments. When using generic continuous space treatments and matching architecture, we observe a 41% improvement upon prior art with cost-effectiveness and 68% improvement upon a similar method in the average treatment effect. The algorithms capture subtle variations in treatment space, structures the efficient optimizations techniques, and opens up the arena for many applications.

Heterogeneous Causal Learning for Effectiveness Optimization in User Marketing

User marketing is a key focus of consumer-based internet companies. Learning algorithms are effective to optimize marketing campaigns which increase user engagement, and facilitates cross-marketing to related products. By attracting users with rewards, marketing methods are effective to boost user activity in the desired products. Rewards incur significant cost that can be off-set by increase in future revenue. Most methodologies rely on churn predictions to prevent losing users to make marketing decisions, which cannot capture up-lift across counterfactual outcomes with business metrics. Other predictive models are capable of estimating heterogeneous treatment effects, but fail to capture the balance of cost versus benefit. We propose a treatment effect optimization methodology for user marketing. This algorithm learns from past experiments and utilizes novel optimization methods to optimize cost efficiency with respect to user selection. The method optimizes decisions using deep learning optimization models to treat and reward users, which is effective in producing cost-effective, impactful marketing campaigns. Our methodology demonstrates superior algorithmic flexibility with integration with deep learning methods and dealing with business constraints. The effectiveness of our model surpasses the quasi-oracle estimation (R-learner) model and causal forests. We also established evaluation metrics that reflect the cost-efficiency and real-world business value. Our proposed constrained and direct optimization algorithms outperform by 24.6% compared with the best performing method in prior art and baseline methods. The methodology is useful in many product scenarios such as optimal treatment allocation and it has been deployed in production world-wide.