SurvCaus : Representation Balancing for Survival Causal Inference
This addresses causal inference for survival data in pathologies, an incremental extension of existing methods to handle censored time-to-event outcomes.
The paper tackled the problem of estimating individual treatment effects for survival outcomes with censorship, proposing a representation balancing framework with theoretical guarantees and demonstrating that it outperforms baseline methods on synthetic and semi-synthetic datasets.
Individual Treatment Effects (ITE) estimation methods have risen in popularity in the last years. Most of the time, individual effects are better presented as Conditional Average Treatment Effects (CATE). Recently, representation balancing techniques have gained considerable momentum in causal inference from observational data, still limited to continuous (and binary) outcomes. However, in numerous pathologies, the outcome of interest is a (possibly censored) survival time. Our paper proposes theoretical guarantees for a representation balancing framework applied to counterfactual inference in a survival setting using a neural network capable of predicting the factual and counterfactual survival functions (and then the CATE), in the presence of censorship, at the individual level. We also present extensive experiments on synthetic and semisynthetic datasets that show that the proposed extensions outperform baseline methods.