English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Counterfactual inference for continuous rather than binary treatmentvariables is more common in real-world causal inference tasks. While there arealready some sample reweighting methods based on Marginal Structural Model foreliminating the confounding bias, they generally focus on removing thetreatment's linear dependence on confounders and rely on the accuracy of theassumed parametric models, which are usually unverifiable. In this paper, wepropose a de-confounding representation learning (DRL) framework forcounterfactual outcome estimation of continuous treatment by generating therepresentations of covariates disentangled with the treatment variables. TheDRL is a non-parametric model that eliminates both linear and nonlineardependence between treatment and covariates. Specifically, we train thecorrelations between the de-confounded representations and the treatmentvariables against the correlations between the covariate representations andthe treatment variables to eliminate confounding bias. Further, acounterfactual inference network is embedded into the framework to make thelearned representations serve both de-confounding and trusted inference.Extensive experiments on synthetic datasets show that the DRL model performssuperiorly in learning de-confounding representations and outperformsstate-of-the-art counterfactual inference models for continuous treatmentvariables. In addition, we apply the DRL model to a real-world medical datasetMIMIC and demonstrate a detailed causal relationship between red cell widthdistribution and mortality.

De-confounding Representation Learning for Counterfactual Inference on Continuous Treatment via Generative Adversarial Network