Disentangle Estimation of Causal Effects from Cross-Silo Data

Estimating causal effects among different events is of great importance to critical fields such as drug development. Nevertheless, Related works Recent years have seen the emergence of machine the data features associated with events may be distributed learning methods for estimating various causal effects across various silos and remain private within respective [11-15]. In single domains, these methods often rely on extensive parties, impeding direct information exchange between experimentation and observations with similar spatial them. This, in turn, can result in biased estimations distribution of data dimensions [16-18]. Inductive approaches of local causal effects, which rely on the characteristics of such as FlextNet [19] leverage structural similarities only a subset of the covariates. To tackle this challenge, we among latent outcomes for causal effect estimation. HTCE introduce an innovative disentangle architecture designed to [20] aids in estimating causal effects in the target domain facilitate the seamless cross-silo transmission of model parameters, with assistance from source domain data, but it is limited to enriched with causal mechanisms, through a combination specific source and target domains. FedCI [21] and Causal-of shared and private branches. Besides, we introduce RFF [22] primarily focus on scenarios where different parties global constraints into the equation to effectively mitigate have the same data feature dimensions.