Balance Regularized Neural Network Models for Causal Effect Estimation
arXiv:2011.11199
Abstract
Estimating individual and average treatment effects from observational data is an important problem in many domains such as healthcare and e-commerce. In this paper, we advocate balance regularization of multi-head neural network architectures. Our work is motivated by representation learning techniques to reduce differences between treated and untreated distributions that potentially arise due to confounding factors. We further regularize the model by encouraging it to predict control outcomes for individuals in the treatment group that are similar to control outcomes in the control group. We empirically study the bias-variance trade-off between different weightings of the regularizers, as well as between inductive and transductive inference.
arXiv commentCausal Discovery & Causality-Inspired Machine Learning Workshop at Neural Information Processing Systems, 2020
References in corpus (11)
- Adam: A Method for Stochastic Optimization
- An Overview of Multi-Task Learning in Deep Neural Networks
- Invariant Risk Minimization
- Causal Effect Inference with Deep Latent-Variable Models
- Domain Adaptation by Using Causal Inference to Predict Invariant Conditional Distributions
- Perfect Match: A Simple Method for Learning Representations For Counterfactual Inference With Neural Networks
- Deep Counterfactual Networks with Propensity-Dropout
- Domain Adaptation as a Problem of Inference on Graphical Models
- Transfer Learning for Estimating Causal Effects using Neural Networks
- MetaCI: Meta-Learning for Causal Inference in a Heterogeneous Population
- Robust Propensity Score Computation Method based on Machine Learning with Label-corrupted Data