Enhancing Counterfactual Classification via Self-Training
This work addresses biased data distributions in applications like pricing and precision medicine, offering an incremental improvement through domain adaptation techniques.
The paper tackles the problem of counterfactual classification with partial feedback, where only outcomes for chosen actions are observed, by proposing a self-training algorithm that imputes pseudolabels to simulate randomized trials. It demonstrates effectiveness on synthetic and real datasets, with concrete performance gains reported in experiments.
Unlike traditional supervised learning, in many settings only partial feedback is available. We may only observe outcomes for the chosen actions, but not the counterfactual outcomes associated with other alternatives. Such settings encompass a wide variety of applications including pricing, online marketing and precision medicine. A key challenge is that observational data are influenced by historical policies deployed in the system, yielding a biased data distribution. We approach this task as a domain adaptation problem and propose a self-training algorithm which imputes outcomes with categorical values for finite unseen actions in the observational data to simulate a randomized trial through pseudolabeling, which we refer to as Counterfactual Self-Training (CST). CST iteratively imputes pseudolabels and retrains the model. In addition, we show input consistency loss can further improve CST performance which is shown in recent theoretical analysis of pseudolabeling. We demonstrate the effectiveness of the proposed algorithms on both synthetic and real datasets.