# Enhancing Counterfactual Classification via Self-Training

arXiv: https://arxiv.org/abs/2112.04461 · HTML: https://mlsanity.com/paper/2112.04461/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 56 references that are in the map.
- First posted on arXiv 8 Dec 2021.

## Builds on (most cited first, 21 of 21)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Virtual Adversarial Training: A Regularization Method for Supervised and Semi-Supervised Learning](https://mlsanity.com/paper/1704.03976/) (2017)
- [Meta-learners for Estimating Heterogeneous Treatment Effects using Machine Learning](https://mlsanity.com/paper/1706.03461/) (2017)
- [Confidence Regularized Self-Training](https://mlsanity.com/paper/1908.09822/) (2019)
- [Learning Representations for Counterfactual Inference](https://mlsanity.com/paper/1605.03661/) (2016)
- [In Defense of Pseudo-Labeling: An Uncertainty-Aware Pseudo-label Selection Framework for Semi-Supervised Learning](https://mlsanity.com/paper/2101.06329/) (2021)
- [Adapting Neural Networks for the Estimation of Treatment Effects](https://mlsanity.com/paper/1906.02120/) (2019)
- [Theoretical Analysis of Self-Training with Deep Networks on Unlabeled Data](https://mlsanity.com/paper/2010.03622/) (2020)
- [Doubly Robust Policy Evaluation and Optimization](https://mlsanity.com/paper/1503.02834/) (2015)
- [Bayesian Inference of Individualized Treatment Effects using Multi-task Gaussian Processes](https://mlsanity.com/paper/1704.02801/) (2017)
- [Confounding-Robust Policy Improvement](https://mlsanity.com/paper/1805.08593/) (2018)
- [Off-policy Bandits with Deficient Support](https://mlsanity.com/paper/2006.09438/) (2020)
- [Human-AI Collaboration with Bandit Feedback](https://mlsanity.com/paper/2105.10614/) (2021)
- [Domain Adaptation for Semantic Segmentation via Class-Balanced Self-Training](https://mlsanity.com/paper/1810.07911/) (2018)
- [A Bayesian Nonparametric Approach for Estimating Individualized Treatment-Response Curves](https://mlsanity.com/paper/1608.05182/) (2016)
- [Unsupervised Domain Adaptation via Calibrating Uncertainties](https://mlsanity.com/paper/1907.11202/) (2019)
- [Cost-Effective Incentive Allocation via Structured Counterfactual Inference](https://mlsanity.com/paper/1902.02495/) (2019)
- [Large-scale Validation of Counterfactual Learning Methods: A Test-Bed](https://mlsanity.com/paper/1612.00367/) (2016)
- [Loss Functions for Discrete Contextual Pricing with Observational Data](https://mlsanity.com/paper/2111.09933/) (2021)
- [Classifying Treatment Responders Under Causal Effect Monotonicity](https://mlsanity.com/paper/1902.05482/) (2019)
- [The Power and Limits of Predictive Approaches to Observational-Data-Driven Optimization](https://mlsanity.com/paper/1605.02347/) (2016)

## Built on by (most cited first, 5 of 5)

- [Probabilistic Conformal Prediction Using Conditional Random Samples](https://mlsanity.com/paper/2206.06584/) (2022)
- [Learning Complementary Policies for Human-AI Teams](https://mlsanity.com/paper/2302.02944/) (2023)
- [Confounding-Robust Policy Improvement with Human-AI Teams](https://mlsanity.com/paper/2310.08824/) (2023)
- [Semi-supervised Batch Learning From Logged Data](https://mlsanity.com/paper/2209.07148/) (2022)
- [Multi-Action Dialog Policy Learning from Logged User Feedback](https://mlsanity.com/paper/2302.13505/) (2023)

## Abstract (arXiv)

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.