Neuroevolutionary Feature Representations for Causal Inference
This work addresses the challenge of improving causal effect estimation for researchers and practitioners in fields like healthcare or economics, though it appears incremental by building on existing neural network approaches.
The paper tackles the problem of estimating heterogeneous treatment effects in causal inference by proposing a novel method that uses a genetic algorithm to learn feature representations optimized for outcome prediction while being less useful for treatment prediction, validated on synthetic and real datasets.
Within the field of causal inference, we consider the problem of estimating heterogeneous treatment effects from data. We propose and validate a novel approach for learning feature representations to aid the estimation of the conditional average treatment effect or CATE. Our method focuses on an intermediate layer in a neural network trained to predict the outcome from the features. In contrast to previous approaches that encourage the distribution of representations to be treatment-invariant, we leverage a genetic algorithm that optimizes over representations useful for predicting the outcome to select those less useful for predicting the treatment. This allows us to retain information within the features useful for predicting outcome even if that information may be related to treatment assignment. We validate our method on synthetic examples and illustrate its use on a real life dataset.