Perfect Match: A Simple Method for Learning Representations For Counterfactual Inference With Neural Networks

Abstract

Learning representations for counterfactual inference from observational data is of high practical relevance for many domains, such as healthcare, public policy and economics. Counterfactual inference enables one to answer "What if...?" questions, such as "What would be the outcome if we gave this patient treatment $t_1$?". However, current methods for training neural networks for counterfactual inference on observational data are either overly complex, limited to settings with only two available treatments, or both. Here, we present Perfect Match (PM), a method for training neural networks for counterfactual inference that is easy to implement, compatible with any architecture, does not add computational complexity or hyperparameters, and extends to any number of treatments. PM is based on the idea of augmenting samples within a minibatch with their propensity-matched nearest neighbours. Our experiments demonstrate that PM outperforms a number of more complex state-of-the-art methods in inferring counterfactual outcomes across several benchmarks, particularly in settings with many treatments.

References in corpus (7)

• Scikit-learn: Machine Learning in Python
• Domain Adaptation: Learning Bounds and Algorithms
• Causal Effect Inference with Deep Latent-Variable Models
• Deep Counterfactual Networks with Propensity-Dropout
• Confounding-Robust Policy Improvement
• A comparison of methods for model selection when estimating individual treatment effects
• Not to Cry Wolf: Distantly Supervised Multitask Learning in Critical Care

Cited by in corpus (20)

• Adapting Neural Networks for the Estimation of Treatment Effects
• DESCN: Deep Entire Space Cross Networks for Individual Treatment Effect Estimation
• A Survey on Causal Inference
• Pareto Invariant Representation Learning for Multimedia Recommendation
• Causal Effect Inference for Structured Treatments
• GeneDisco: A Benchmark for Experimental Design in Drug Discovery
• Learning Individual Causal Effects from Networked Observational Data
• NCoRE: Neural Counterfactual Representation Learning for Combinations of Treatments
• Adapting Neural Networks for Uplift Models
• Counterfactual Evaluation of Treatment Assignment Functions with Networked Observational Data
• A Causal Lens for Peeking into Black Box Predictive Models: Predictive Model Interpretation via Causal Attribution
• MultiMBNN: Matched and Balanced Causal Inference with Neural Networks
• Balance Regularized Neural Network Models for Causal Effect Estimation
• Regret Minimization for Causal Inference on Large Treatment Space
• GraphITE: Estimating Individual Effects of Graph-structured Treatments
• Combining Incomplete Observational and Randomized Data for Heterogeneous Treatment Effects
• VACA: Design of Variational Graph Autoencoders for Interventional and Counterfactual Queries
• Estimating Potential Outcome Distributions with Collaborating Causal Networks
• Hi-CI: Deep Causal Inference in High Dimensions
• Adaptive Multi-Source Causal Inference