Decision Focused Causal Learning for Direct Counterfactual Marketing Optimization
Hao Zhou, Rongxiao Huang, Shaoming Li, Guibin Jiang, Jiaqi Zheng, Bing Cheng, Wei Lin
TL;DR
The paper addresses direct counterfactual marketing optimization under budget uncertainty, formalized as MTBAP with $F(z,B)=\sum_i\sum_j z_{ij} r_{ij}$ and constraints $\sum_i\sum_j z_{ij} c_{ij} \le B$, $z_{ij}\in\{0,1\}$. It introduces Decision-Focused Causal Learning (DFCL), which uses Lagrangian duality to convert uncertain constraints into a dual objective and trains models with a dual decision loss $\mathcal{L}_{DDL}$, together with surrogate losses such as $\mathcal{L}_{PLL}$ (policy learning) and $\mathcal{L}_{MERL}$ (maximum entropy regularized). To handle counterfactuals and large-scale data, the framework adopts an improved finite-difference gradient estimator guided by Expected Outcome Metric (EOM), and a scalable learning framework combining prediction and decision objectives. Empirical results on offline datasets and online A/B tests (including deployment in Meituan) show that DFCL outperforms state-of-the-art baselines in decision quality (AUCC, EOM) and achieves meaningful revenue gains, demonstrating practical impact for budget-aware marketing campaigns.
Abstract
Marketing optimization plays an important role to enhance user engagement in online Internet platforms. Existing studies usually formulate this problem as a budget allocation problem and solve it by utilizing two fully decoupled stages, i.e., machine learning (ML) and operation research (OR). However, the learning objective in ML does not take account of the downstream optimization task in OR, which causes that the prediction accuracy in ML may be not positively related to the decision quality. Decision Focused Learning (DFL) integrates ML and OR into an end-to-end framework, which takes the objective of the downstream task as the decision loss function and guarantees the consistency of the optimization direction between ML and OR. However, deploying DFL in marketing is non-trivial due to multiple technological challenges. Firstly, the budget allocation problem in marketing is a 0-1 integer stochastic programming problem and the budget is uncertain and fluctuates a lot in real-world settings, which is beyond the general problem background in DFL. Secondly, the counterfactual in marketing causes that the decision loss cannot be directly computed and the optimal solution can never be obtained, both of which disable the common gradient-estimation approaches in DFL. Thirdly, the OR solver is called frequently to compute the decision loss during model training in DFL, which produces huge computational cost and cannot support large-scale training data. In this paper, we propose a decision focused causal learning framework (DFCL) for direct counterfactual marketing optimization, which overcomes the above technological challenges. Both offline experiments and online A/B testing demonstrate the effectiveness of DFCL over the state-of-the-art methods. Currently, DFCL has been deployed in several marketing scenarios in Meituan, one of the largest online food delivery platform in the world.