Mechanisms for Data Sharing in Collaborative Causal Inference (Extended Version)

TL;DR

This paper introduces a data valuation scheme for collaborative causal inference that incentivizes data sharing among self-interested parties by fairly rewarding contributions based on their impact on causal structure learning.

Contribution

It proposes a novel evaluation scheme tailored for causal inference to measure data contributions and designs mechanisms to incentivize data sharing in federated settings.

Findings

The valuation scheme effectively quantifies each party's data contribution.

Mechanisms based on the scheme promote fair data sharing.

The approach enhances collaborative causal inference accuracy.

Abstract

Collaborative causal inference (CCI) is a federated learning method for pooling data from multiple, often self-interested, parties, to achieve a common learning goal over causal structures, e.g. estimation and optimization of treatment variables in a medical setting. Since obtaining data can be costly for the participants and sharing unique data poses the risk of losing competitive advantages, motivating the participation of all parties through equitable rewards and incentives is necessary. This paper devises an evaluation scheme to measure the value of each party's data contribution to the common learning task, tailored to causal inference's statistical demands, by comparing completed partially directed acyclic graphs (CPDAGs) inferred from observational data contributed by the participants. The Data Valuation Scheme thus obtained can then be used to introduce mechanisms that incentivize the agents to contribute data. It can be leveraged to reward agents fairly, according to the quality of their data, or to maximize all agents' data contributions.