🤖 AI Summary
In vertical federated learning (VFL), securely and fairly evaluating feature contributions remains challenging under model-agnostic assumptions, especially without access to raw data or shared model training.
Method: This paper proposes the first privacy-preserving Shapley-CMI framework, integrating Shapley values with conditional mutual information (CMI) to enable model-agnostic, data- and training-free feature valuation. We design an encrypted permutation and intersection-size computation protocol based on private set intersection (PSI) servers, achieving fully decentralized, trustless implementation. To enhance efficiency, we introduce ID-based grouping discretization and lightweight homomorphic operations.
Contributions/Results: Extensive experiments validate the framework’s correctness, end-to-end privacy guarantees, and scalability across multi-party VFL settings. Our approach establishes an auditable, interpretable foundation for fair feature valuation in early-stage VFL data collaboration—without requiring raw data sharing, model exchange, or a trusted third party.
📝 Abstract
Federated Learning (FL) is an emerging machine learning paradigm that enables multiple parties to collaboratively train models without sharing raw data, ensuring data privacy. In Vertical FL (VFL), where each party holds different features for the same users, a key challenge is to evaluate the feature contribution of each party before any model is trained, particularly in the early stages when no model exists. To address this, the Shapley-CMI method was recently proposed as a model-free, information-theoretic approach to feature valuation using Conditional Mutual Information (CMI). However, its original formulation did not provide a practical implementation capable of computing the required permutations and intersections securely. This paper presents a novel privacy-preserving implementation of Shapley-CMI for VFL. Our system introduces a private set intersection (PSI) server that performs all necessary feature permutations and computes encrypted intersection sizes across discretized and encrypted ID groups, without the need for raw data exchange. Each party then uses these intersection results to compute Shapley-CMI values, computing the marginal utility of their features. Initial experiments confirm the correctness and privacy of the proposed system, demonstrating its viability for secure and efficient feature contribution estimation in VFL. This approach ensures data confidentiality, scales across multiple parties, and enables fair data valuation without requiring the sharing of raw data or training models.