About this document

Adversarial De-confounding for ITE Estimation by s711333112 is a document available to read on EtoBox.

This paper addresses the challenge of estimating individualised treatment effects (ITE) from observational data, focusing on the issue of de-confounding. It proposes a novel approach using disentangled representations with adversarial training to balance confounders in binary treatment settings, improving the accuracy of ITE estimation compared to existing methods. Empirical results demonstrate that this approach effectively reduces error in ITE estimation across various datasets.

Author

s711333112

Language

EN