Towards Privacy-Aware Causal Structure Learning in Federated Setting
This work addresses privacy concerns in causal structure learning for applications where data cannot be centralized, though it appears incremental as it adapts existing methods to a federated context.
The paper tackles the problem of causal structure learning in privacy-preserving federated settings by proposing a novel Federated PC (FedPC) algorithm with layer-wise aggregation and consistent separation set strategies, and validates its effectiveness through extensive experiments.
Causal structure learning has been extensively studied and widely used in machine learning and various applications. To achieve an ideal performance, existing causal structure learning algorithms often need to centralize a large amount of data from multiple data sources. However, in the privacy-preserving setting, it is impossible to centralize data from all sources and put them together as a single dataset. To preserve data privacy, federated learning as a new learning paradigm has attracted much attention in machine learning in recent years. In this paper, we study a privacy-aware causal structure learning problem in the federated setting and propose a novel Federated PC (FedPC) algorithm with two new strategies for preserving data privacy without centralizing data. Specifically, we first propose a novel layer-wise aggregation strategy for a seamless adaptation of the PC algorithm into the federated learning paradigm for federated skeleton learning, then we design an effective strategy for learning consistent separation sets for federated edge orientation. The extensive experiments validate that FedPC is effective for causal structure learning in a federated learning setting.