论文标题
因果推断的神经进化特征表示
Neuroevolutionary Feature Representations for Causal Inference
论文作者
论文摘要
在因果推论的领域内，我们考虑了估计数据中异质治疗效应的问题。我们建议并验证一种学习特征表征的新方法，以帮助估计条件平均治疗效果或CATE。我们的方法着重于训练有素的神经网络中的中间层，以预测特征的结果。与以前的方法相比，鼓励表示表示为治疗不变的方法，我们利用一种遗传算法优化而不是预测结果的代表性，以选择那些对预测治疗的不太有用的算法。这使我们可以将信息保留在对预测结果有用的功能中，即使该信息可能与治疗分配有关。我们验证了合成示例的方法，并说明了其在现实生活数据集上的使用。
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.
