Deep-Treat: Learning Optimal Personalized Treatments From Observational Data Using Neural Networks

By Onur Atan, James Jordon, Mihaela van der Schaar. 2018

76 citations recorded by the source corpus.

We propose a novel approach for constructing effective treatment policies when the observed data is biased and lacks counterfactual information. Learning in settings where the observed data does not contain all possible outcomes for all treatments is difficult since the observed data is typically biased due to existing clinical guidelines. This is an important problem in the medical domain as collecting unbiased data is expensive and so learning from the wealth of existing biased data is a worthwhile task. Our approach separates the problem into two stages: first we reduce the bias by learning a representation map using a novel auto-encoder network---this allows us to control the trade-off between the bias-reduction and the information loss---and then we construct effective treatment policies on the transformed data using a novel feedforward network. Separation of the problem into these two stages creates an algorithm that can be adapted to the problem at hand---the bias-reduction step can be performed as a preprocessing step for other algorithms. We compare our algorithm against state-of-art algorithms on two semi-synthetic datasets and demonstrate that our algorithm achieves a significant improvement in performance.

Subfields: Artificial Intelligence, Statistics and Probability, Causal Inference & Discovery, Medical & Health AI

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Statistics and Probability

In Causal Inference & Discovery

In Medical & Health AI

Citations in the corpus

Towards Causal Representation Learning (2021, background): “While it may be useful to automate certain decisions, an understanding of causality may be necessary to recommend treatment options that are personalized and reliable [201, 242, 224, 273, 6, 3, 30, 165].”
Limits of Estimating Heterogeneous Treatment Effects: Guidelines for Practical Algorithm Design (2018, result): “Similar observations are reported in (Alaa & van der Schaar, 2017; Atan et al., 2018), where the selection of the model structure seemed to influence the achieved CATE estimation performance even when selection bias is not accounted for.”
A Survey on Causal Inference (2020, methodology): “1]. For the direction of utilizing observational dataset, various methods derive the personalized treatment rules guided by the estimated ITE under the unconfoundedness assumption, such as deep-treat [7], tiered case-cohort design based method [66]. However, in this area, there are limited works to ”
Bayesian Nonparametric Causal Inference: Information Rates and Learning Algorithms (2017, background): “Instead, most previous works on causal effect inference have focused on model development, and various algorithms have been recently developed for estimating individualized treatment effects from observational data, mostly based on either treebased methods [7], [8], [16], or deep learning methods [1”